Professional Documents
Culture Documents
Guide To Instruments and Methods of Observation: Volume IV - Space-Based Observations
Guide To Instruments and Methods of Observation: Volume IV - Space-Based Observations
Observation
Volume IV – Space-based Observations
2018 edition
WEATHER CLIMATE WATER
WMO-No. 8
Guide to Instruments and Methods of
Observation
Volume IV – Space-based Observations
2018 edition
WMO-No. 8
EDITORIAL NOTE
METEOTERM, the WMO terminology database, may be consulted at http://public.wmo.int/en/
resources/meteoterm.
Readers who copy hyperlinks by selecting them in the text should be aware that additional
spaces may appear immediately following http://, https://, ftp://, mailto:, and after slashes (/),
dashes (-), periods (.) and unbroken sequences of characters (letters and numbers). These spaces
should be removed from the pasted URL. The correct URL is displayed when hovering over the
link or when clicking on the link and then copying it from the browser.
WMO-No. 8
The right of publication in print, electronic and any other form and in any language is reserved by
WMO. Short extracts from WMO publications may be reproduced without authorization, provided
that the complete source is clearly indicated. Editorial correspondence and requests to publish,
reproduce or translate this publication in part or in whole should be addressed to:
ISBN 978-92-63-10008-5
NOTE
The designations employed in WMO publications and the presentation of material in this publication do
not imply the expression of any opinion whatsoever on the part of WMO concerning the legal status of any
country, territory, city or area, or of its authorities, or concerning the delimitation of its frontiers or boundaries.
The mention of specific companies or products does not imply that they are endorsed or recommended by
WMO in preference to others of a similar nature which are not mentioned or advertised.
PUBLICATION REVISION TRACK RECORD
Part/
Date chapter/ Purpose of amendment Proposed by Approved by
section
CONTENTS
Page
CHAPTER 1. INTRODUCTION. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.1 Historical perspective. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.2 Spatial and temporal scales . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.3 Complementary nature of space-based and surface-based measurements . . . . . . . . . . . 2
Page
Page
Page
Page
On 1 April 1960, a new era started for meteorology with the launch of the Television and Infrared
Observation Satellite – 1 (TIROS‑1). Weather systems, which had only been depicted until then
by synoptic maps and aircraft observations, could be visualized at a glance. Their rapidly evolving
nature became more evident with geostationary imagery from the Applications Technology
Satellite – 1 (ATS‑1), launched on 6 December 1966. The term “nowcasting” emerged, becoming
the first application of meteorological satellites.
Initially, satellite data were nearly exclusively used for nowcasting. They were first applied to the
field of numerical weather prediction, starting with Nimbus‑3 (13 April 1969), by using data from
experimental instruments to derive vertical profiles of atmospheric temperature and humidity
and by deriving cloud-motion winds from geostationary image sequences.
The First Global Atmospheric Research Programme (GARP) Global Experiment (FGGE, 1979–
1980) was able to assemble, for the first time, a composite system of four geostationary satellites
and two near-polar satellites, which delivered global sounding and imaging coverage four times
a day and imagery at low and mid latitudes every half hour. It is important to note that since their
early days, in addition to supporting operational applications, meteorological satellites have
enabled advances in the understanding of atmospheric dynamics and climate.
Driven by the high economic value of earth resource exploration and vegetation cycle
monitoring, new satellite programmes emerged with a focus on land surface observation.
Landsat‑1, launched on 23 July 1972, led the first series of high-resolution land observation
satellites, and coverage from the Satellite pour l’Observation de la Terre (SPOT) series, beginning
on 22 February 1986 with SPOT‑1, provided imagery at a spatial resolution of 10 to 20 metres.
Exploration of the ocean began with the launch of SeaSat on 27 June 1978, which marked the
advent of all-weather microwave sensing, both active and passive. Almost simultaneously, on
24 October 1978, Nimbus‑7 used passive microwave sensing with the addition of ocean colour
monitoring. After the SeaSat altimetry, scatterometry and synthetic aperture radar imagery
missions, no active sensing mission was operated until the launch of the European Remote-
sensing Satellite – 1 (ERS‑1) on 17 July 1991. The retrieval of information on atmospheric radiation
and chemistry was initially explored by several Nimbus missions. A milestone for Earth radiation
study was the Earth Radiation Budget Satellite (ERBS), launched on 5 October 1984. For
atmospheric chemistry, a major milestone was the Upper Atmosphere Research Satellite (UARS),
launched on 12 September 1991.
The concept of the Global Observing System (GOS) was totally revised with the advent of
satellites, taking advantage of the complementary nature of surface-based and space-based
observations. The space-based component offers the unique opportunity of uninterrupted
global coverage and frequent observing cycles. A striking advantage is the capability of vertical
atmospheric sounding over the oceans, alleviating a great limitation of observations for global
numerical weather prediction. Over continental areas, observing networks are biased towards
populated areas, whereas the vast majority of land surfaces are relatively unpopulated and hence
undersampled; furthermore, some local observations available from the ground (for example,
cloud type) are hard to integrate spatially.
One important difference between satellite and surface measurements is the integration in
space and time. Satellite measurements integrate the incoming signal over an instantaneous
field of view determined by the need to collect sufficient radiant energy to provide the required
2 GUIDE TO INSTRUMENTS AND METHODS OF OBSERVATION - VOLUME IV
It is acknowledged that satellites are unable to perform all needed observations with the
required measurement quality. For certain geophysical variables, no remote-sensing principle
exists. For others, the required measurement quality is only achievable with ancillary information
from accurate surface-based observing systems. In addition, since satellite measurements
are often of indirect nature (the primary observed quantity being radiation), surface-based
measurements play a key role for the validation of satellite-derived products.
There are still areas where exclusively surface-based systems can provide measurements of
acceptable quality. However, even in those cases, satellites can be useful in spatially extending
local and sparse ground measurements. In particular, the practice of assimilation makes
it possible to transfer information across geophysical variables measured using different
techniques: this means that satellite observations may contribute to the knowledge of
geophysical variables even when not directly observed from satellite, provided that there is a
strong physical relationship between these variables. Synergistic use of surface-based and space-
based observations is fundamental to the WMO Integrated Global Observing System.
Note: Detailed descriptions of satellite programmes and instruments are available in the WMO online database of
space-based observation capabilities: http://w ww.wmo.int/oscar.
CHAPTER 2. PRINCIPLES OF EARTH OBSERVATION FROM SPACE
This chapter provides an overview of Earth observation from space, including its potential
benefits and limitations. It describes basic concepts of orbits and the characteristics of Earth
viewing from space. It also introduces the principles of remote-sensing.
The Earth can be observed from space from different orbits under various viewing conditions.
The following issues are considered in this section:
(b) Orbital period, geostationary orbit, observing cycle and repeat cycle;
The greatest advantage of observing from a satellite platform, rather than from the ground or a
balloon, is the wide potential field of view (FOV). Satellite observing platforms usually orbit at
a minimum height of 400 km. Often, they orbit much higher, some as far as the geostationary
orbit (35 786 km). The FOV depends on the orbital height, the instrument configuration and the
intended application. Those may limit the useful range of zenith angles (ζ) under which the Earth
can be viewed. If the satellite FOV is characterized as the maximum ground distance potentially
viewed from satellite height under a given zenith angle, the relationship set out in Figure 2.1 can
be expressed as:
R
FOV = 2 R δ π 180 sin (ζ − δ ) = sin ζ (2.1)
H +R
where R = 6 371 km (Earth’s radius), H = orbital height in km, and δ = geocentric angle in degrees.
Table 2.1 presents values of the satellite FOV (in km) as a function of orbital height for typical
values of zenith angle ζ. The corresponding geocentric angle δ is also shown.
The potential satellite FOV may not be entirely covered by a single instrument. Either the
sensing principle or the technological features of an instrument may set an upper limit to its
FOV. For instance, radar altimeters can only operate in a nadir geometry. They therefore have
H
ζ
R δ
Table 2.1. Potential satellite field of view and corresponding geocentric angle, as functions of
satellite height and zenith angle, under which the Earth’s spot is viewed
ζ = 85°
3 423 km 15.39° 4 322 km 19.43° 5 057 km 22.74° 16 978 km 76.34°
(telecommunications)
ζ = 70°
1 746 km 7.85° 2 405 km 10.82° 2 980 km 13.40° 13 752 km 61.84°
(qualitative use)
ζ = 60°
1 207 km 5.43° 1 707 km 7.68° 2 157 km 9.70° 11 671 km 52.48°
(quantitative use)
no proper FOV, except for the broadening of the beam due to diffraction. Very high-resolution
imagers usually have an FOV within a range of several tens of kilometres, as do synthetic aperture
radars (SARs).
The satellite motion enables the instrument to cover successive FOVs along the orbit. These
constitute a strip of observed Earth surface called a swath. The swath may be centred along the
sub-satellite track, or be parallel to it for side-looking instruments (for example, SARs). For several
purposes (steerable pointing for emergencies, stereoscopy in association with successive orbits,
and the like), certain instruments with limited swath may tilt the swath to the side of the track
within what is called a field of regard. The swath width is a cross-track component of the actual
FOV of the instrument. The swath is not defined for instruments in geostationary orbit.
2.1.2 Orbital period, geostationary orbit, observing cycle and repeat cycle
The orbital height H determines the orbital period T. The relationship is:
3
H 2
T = a (1 + ) (2.2)
R
where a = 84.47 min (T resulting in minutes).
The height, which corresponds to one sidereal day (23 h 56 min 04 s) is 35 786 km. A satellite
orbiting at this height is called geosynchronous. The orbit is called geostationary if the orbit lies
in the equatorial plane and is run eastward: the satellite appears steady compared to the Earth’s
surface on the nadir of the equatorial sub-satellite point.
For an inclined orbit with respect to the equatorial plane, the satellite will cross the Equator at
a certain longitude. After T minutes, there will be another equatorial crossing at a longitude
displaced westward by the number of degrees that corresponds to the Earth’s rotation during the
orbital period. The difference of longitude (or space distance) between two successive equatorial
crossings in the same phase (descending or ascending) is called decalage. Together with the
instrument swath, decalage determines the time needed for a full Earth surface observation
(observing coverage) (Figure 2.2).
If the instrument swath is at least as large as the decalage, the coverage provided by the two
contiguous orbits is continuous. Therefore, the time needed for global coverage (observing
cycle) depends on the ratio between instrument swath and decalage.
Table 2.2 shows the period and corresponding decalage for the orbital height examples
in Table 2.1. In addition, observing cycles corresponding to several instrument swaths are
presented. Those swaths are associated with qualitative and quantitative use; that is with 70° and
CHAPTER 2. PRINCIPLES OF EARTH OBSERVATION FROM SPACE 5
Swath
Decalage
Figure 2.2. Decalage between two successive orbits, and instrument swath
60° zenith angles, respectively. Observing time is halved for instruments capable of observation
during the day and night. The decalage and observing cycle are not quoted for H = 35 786 km
(geostationary altitude).
For a geostationary satellite, which continuously views the same area of the Earth’s surface, the
observing cycle is only determined by instrument characteristics and may take a few minutes
or less to complete, depending on the area scanned. Within the satellite’s area of coverage,
geostationary observation is perfectly suited for continuous monitoring. For example, such
monitoring is needed to detect instantaneous events like lightning strokes, or for high-frequency
temporal sampling of rapidly evolving situations, such as active convection. However, the
coverage excludes very high latitudes or any locations too far from the sub-satellite point.
Table 2.1 shows that δ = 81.31° is the maximum geocentric angle.
For a non-geostationary satellite, the orbit is said to have a repeat cycle if it overpasses the
same track exactly after a given number of revolutions. During the timespan of a repeat cycle,
the satellite track may shift from day to day, following a determined pattern that may exhibit
certain periodicities called sub-cycles. Some sub-cycles may be of interest because distinct areas,
relatively close to each other, are visited within short time intervals; other sub-cycles may be of
interest because the areas covered are spatially adjacent.
If the orbit is Sun-synchronous (see section 2.1.3), the existence of a repeat cycle means that a
whole number of revolutions can be completed in exactly a whole number of days. The orbital
period determines N, the number of orbits that the satellite runs in 24 h. This is normally not an
integer. In order to obtain a repeat cycle of m days, the orbital period is adjusted to ensure that
N multiplied by m is an integer. N can then be expressed in the following form, where n and ℓ
are, respectively, the quotient and remainder of the integer division of “N ∙ m” by m:
N = n + m (2.3)
where n, ℓ and m are integers (ℓ < m).
Table 2.2. Period, decalage and observing cycle for the orbits indicated in Table 2.1
Table 2.3. Repeat cycles and main sub-cycles for a number of orbits
Non-Sun-
Sun-synchronous orbits synchronous
orbit
909 km 705 km 820 km
832 km 791 km 1 336 km
Orbital height (e.g. Landsat (e.g. Landsat (e.g.
(e.g. SPOT) (e.g. Envisat) (e.g. JASON)a
1–3) 4–8) Metop)
Period 103.2 min 98.9 min 101.5 min 100.6 min 101.3 min 112.4 min
Main sub-cycle(s) 1 day 7 days, 2 days 5 days 16 days, 3 days 5 days 3 daysa
Note:
a In the case of the Joint Altimetry Satellite Oceanography Network (JASON), which is not Sun-synchronous, the
figures refer to a day of 23 h 48 min in duration, that is, 0.99156 of the duration of a solar day.
Equation 2.3 also applies to non-Sun-synchronous orbits. However, in such cases, the repeat
cycle m is no longer expressed in solar days of 24 h but must account for a slight correction due
to the drift of the orbit. Table 2.3 provides examples of repeat cycles and main sub-cycles for a
number of orbits.
Decalage
Strip Strip Strip Strip Strip Strip Strip Strip Strip Strip Strip Strip Strip Strip Strip Strip Strip Strip Strip
19 18 17 16 15 14 13 12 11 10 09 08 07 06 05 04 03 02 01
Orbit Orbit Day
02 01 01
Orbit Day
15 02
1 day sub-cycle Orbit Day
29 03
Orbit Day
43 04
Orbit Day
57 05
Orbit Day
71 06
Orbit Day
85 07
Orbit Day
99 08
Orbit Day
113 09
Orbit Day
127 10
Orbit Day
141 11
Orbit Day
155 12
Orbit Day
169 13
Orbit Day
183 14
Orbit Day
197 15
Orbit Day
211 16
Orbit Day
225 17
Orbit Day
239 18
Orbit Orbit Day
253 252 19
Figure 2.3. Schematic evolution of the orbital track of early Landsat over a repeat cycle
(N = 13 + 17/18, repeat cycle: 18 days, 251 revolutions/cycle)
CHAPTER 2. PRINCIPLES OF EARTH OBSERVATION FROM SPACE 7
An orbit with a repeat cycle is a necessary feature if a certain location needs to be viewed at
fixed intervals under identical conditions. This is true of altimetric measurements for geodetic
application, or of high-resolution land observation imagers, used to detect local variations.
Repeat cycles may be useful when the instrument swath is substantially narrower than the
decalage and global coverage cannot be achieved in a single day. In that case, the sequence of
coverage over successive days can be arranged to follow a certain logic if requested. That logic
might be to provide regular progression, or to avoid biases due to unsuitable sampling.
Figure 2.3 shows the pattern evolution of orbital passes for an orbit with a one-day sub-cycle
(such as for early Landsat). As Figure 2.3 shows, the one-day sub-cycle ensures that each day,
the covered strip is adjacent to the one that was observed on the previous day. The width of the
covered strip can be tuned to the instrument swath so as to avoid any gaps. The drawback is that,
after the first few daily visits over or close to the target area, the next sequence of visits occurs
only after the completion of the repeat cycle.
With the current Landsat, the temporal evolution of the orbit tracks during the repeat cycle
(16 days) results in two main sub-cycles, as shown in Figure 2.4. The two-day sub-cycle ensures a
shorter temporal gap, but the seven-day sub-cycle provides a closer geographical match.
Although the concept of repeat cycles and sub-cycles stems from the requirements placed on the
use of narrow-swath instruments, including those with nadir-only viewing, orbits with sub-cycles
may also be useful for relatively wide-swath instruments. For example, sounding instruments
may have a swath as wide as several thousand kilometres. (For example, the Advanced
Decalage
Strip Strip Strip Strip Strip Strip Strip Strip Strip Strip Strip Strip Strip Strip Strip Strip Strip
17 16 15 14 13 12 11 10 09 08 07 06 05 04 03 02 01
Orbit Orbit Day
02 01 01
2 day sub-cycle Orbit Day
16 02
Orbit Day
31 03
Orbit Day
45 04
Orbit Day
60 05
Orbit Day
74 06
Orbit Day
7 day sub-cycle 89 07
Orbit Day
103 08
Orbit Day
118 09
Orbit Day
133 10
Orbit Day
147 11
Orbit 9 day sub-cycle Day
162 12
Orbit Day
176 13
Orbit Day
191 14
Orbit Day
205 15
Orbit Day
220 16
Orbit Orbit Day
235 234 17
Figure 2.4. Schematic evolution of the orbital track of current Landsat over a repeat cycle
(N = 14 + 9/16, repeat cycle: 16 days, 233 revolutions/cycle). Three sub-cycles are shown:
seven days (westbound), the main one, providing the closest observations in space; two days
(eastbound), for closer observations in time; and nine days (eastbound), of marginal interest,
the sum of the first two sub-cycles.
8 GUIDE TO INSTRUMENTS AND METHODS OF OBSERVATION - VOLUME IV
Decalage
30 29 28 27 26 25 24 23 22 21 20 19 18 17 16 15 14 13 12 11 10 09 08 07 06 05 04 03 02 01
02 01
02 16 02
03 30 03
04 44 04
05 58 05
06 73 06
07 87 07
08 101 08
09 115 09
10 129 10
11 144 11
12 158 12
13 172 13
5 day sub-cycle
14 186 14
15 4 day sub-cycle 200 15
16 215 16
17 229 17
18 243 18
19 257 19
20 271 20
21 286 21
22 300 22
23 314 23
24 328 24
25 342 25
26 357 26
27 351 27
28 385 28
29 399 29
414 413
Figure 2.5. Schematic evolution of the orbital track of Metop (N = 14 + 6/29, repeat cycle:
29 days, 412 revolutions/cycle). Two sub-cycles are shown: five days (eastbound), the main
one, providing the closest observations in space; and four days (westbound), for marginally
closer observations in time.
Microwave Sounding Unit (AMSU) or the Infrared Atmospheric Sounding Interferometer (IASI)
have swaths of over 2 200 km). However, the quality of the products retrieved is higher when
closer to the nadir sub-track. Therefore, there is an interest in ensuring that the coverage provides
a fair blend of higher and lower quality data. Orbits for the National Oceanic and Atmospheric
Administration (NOAA) satellites and Meteorological Operational (Metop) satellites have a five-
day sub-cycle, as shown in Figure 2.5 for Metop.
Repeat cycles and sub-cycles are convenient for several reasons, but a few drawbacks should also
be noted:
(a) If the instrument swath is too narrow with respect to the decalage and the number of
orbital passes during the repeat cycle, some areas will never be observed. An extreme case
is a nadir-only viewing instrument such as an altimeter.
(b) The day-to-day sequence of observations from a repeat-cycle orbit may introduce sampling
biases in the observations (a spurious wavelength corresponding to a repeat cycle or
sub-cycles).
(c) Maintenance of the repeat cycle/sub-cycles requires costly satellite orbit control systems.
Therefore, if all instruments on board have a sufficiently wide swath, a repeat cycle or sub-cycle is
generally not carried out.
CHAPTER 2. PRINCIPLES OF EARTH OBSERVATION FROM SPACE 9
North Pole
Equatorial
plane
Orbital plane
Altitude
The orbital plane can lie in the Earth’s equatorial plane or be inclined by ε degrees (see
Figure 2.6).
For ε = 90°, the satellite follows a meridian line and the orbit is polar. This is very convenient for
observing the Earth’s surface pole-to-pole.
The gravity field acting on the satellite is perpendicular to the geopotential surface at satellite
altitude, which is slightly ellipsoid like the geoid. Where ε ≠ 90°, the effect of these forces is a
precession of the orbital plane around the polar axis. The precession rate α is computed as:
7
H −2
α = −10.02 cos ε (1 + ) ( degree day ) (2.4)
R
For a purely polar orbit (ε = 90°), the precession rate is thus zero. The orbital plane has an
invariant orientation with respect to the fixed stars. However, as the Earth rotates around
the Sun over one year, the illumination conditions of the surface, as viewed by the satellite,
change every day by 360/365 degrees; that is 59 min. An area viewed in daylight at noon on
day t0 will be viewed in dawn conditions on day t = t0 + 3 months (Figure 2.7, left panel). For
measurements in daylight, this would mean different observing conditions day after day, with
seasonally-dependent observation times. In particular, when the Earth–Sun direction becomes
perpendicular to the orbital plane, the illumination in dawn–dusk conditions makes many
measurements impossible.
The orbital inclination ε can be set in such a way that the precession rate exactly matches the
yearly revolution of the Earth around the Sun. By imposing the value α = 360/365 (degree/day) in
equation 2.4, it is found that the inclination ε 0 must satisfy:
7
H
cos ε 0 = −0.0988 (1 + ) 2 (2.5)
R
An orbit that satisfies that condition is called Sun-synchronous. The negative value of ε 0 indicates
that the orbit is retrograde with respect to the Earth’s rotation. The local solar time of the areas
overflown by the satellite at a given latitude is constant across the whole year (see Figure 2.7,
right panel). Table 2.4 presents the ε 0 values of a number of Sun-synchronous orbits as a function
of orbital height.
Note that the deviation from the polar axis increases with orbital height. This is a drawback for
high Sun-synchronous orbits: the poles might not be observed unless the instrument swath is
wide enough. However, for relatively low orbital heights, the orbital plane is near-polar.
The most important feature of a Sun-synchronous orbit – the fixed local solar time – may be a
disadvantage for certain types of measurements. Diurnally-varying phenomena (for example,
convective clouds, precipitation, radiation budget, sea level affected by astronomical tides)
display biased sampling if observed from a Sun-synchronous satellite (that is, at a fixed local solar
time).
In general, satellites for operational meteorology, land observation and oceanography, with the
exception of geodetic-quality altimetry, use a Sun-synchronous orbit. Scientific missions focused
on processes affected by diurnal variation, which require unbiased sampling, may favour non-
Sun-synchronous (drifting) orbits.
The previous sections are applicable to circular orbits, which are by far the most widely used in
Earth observation, particularly for Sun-synchronous and geostationary orbits. However, both
near-polar low Earth orbits (LEO) and geostationary Earth orbits (GEO) have several limitations.
A near-polar LEO satellite provides global but infrequent coverage. Even if the instrument
swath is as large as the decalage, thus providing contiguous coverage by consecutive orbits,
one satellite can cover the whole of the Earth’s surface twice a day at most (or even once a day
if sensing can be performed either in daylight only or only in night-time conditions). If more
frequent global coverage is needed, additional satellites in complementary orbital planes are
necessary (see Table 2.5).
It is clear from Table 2.5 that any observing cycle shorter than, for instance, three hours, would
be extremely demanding, as it would require a constellation of LEO satellites in coordinated
orbits.
Table 2.5. Number of LEO satellites needed to achieve a required observing cycle
(assumed height: H = 800 km)
Winter Winter
18/06 LST 18/06 LST
Polar orbit Sun-synchronous orbit
Figure 2.7. Left: Pure polar orbit with changing local solar time (LST) throughout the year;
right: Sun-synchronous orbit with fixed LST throughout the year
CHAPTER 2. PRINCIPLES OF EARTH OBSERVATION FROM SPACE 11
The coverage improves substantially for high latitudes (see Figure 2.2). For example, coverage is
twice as frequent at 60° latitude than at the Equator. In polar regions, the frequency of coverage
becomes close to the orbital period T (~100 min) or sub-hourly with more satellites.
A shorter observing cycle may be obtained by leaving aside the Sun-synchronous feature and
adopting a lower inclination, but the coverage is then no longer global. Low-inclination orbits
are used for monitoring the intertropical zones.
The GEO orbit provides observations at a rate that is limited only by the instrument. However, a
constellation of about six satellites around the Equator is needed to cover all longitude sectors up
to a latitude of at least 55°; the highest latitudes cannot be covered.
Some of these limitations can be mitigated by adopting an elliptical orbit. On an elliptical orbit,
the satellite speed changes along the orbit; it is minimal around the apogee allowing more time
for acquiring measurements from the overflown area. Elliptical orbits are usually optimized for
specific purposes, particularly for space science, such as to collect in situ measurements up to
very high altitudes by physically passing through the ionosphere and plasmasphere.
One problem of elliptical orbits is that, since the argument of the perigee is affected by the
secular perturbation, the apogee occurs at latitudes that change with time. The secular
perturbation can be compensated for if the orbital inclination is ε = sin–1 (4/5)1/2 ≈ 63.4°. In this
case, the apogee region where the satellite dwells for most of the time is stable. In that position,
measurements can be taken very frequently, in a quasi-geostationary fashion.
Two orbits of this kind have been used for telecommunication satellites and are planned to be
used for Earth observation: Molniya (Figure 2.8), which has a 12 h period and an apogee at
39 800 km; and Tundra, which has a 24 h period and an apogee at 48 300 km. In the Molniya
orbit, the satellite is nearly geostationary for about 8 h of the 12 h period. In the Tundra orbit, it is
nearly geostationary for about 16 h of the 24 h period.
The Molniya and Tundra orbits only serve one hemisphere. In addition, the 8 h or 12 h quasi-
geostationary observing area is centred on a specific local solar time. If all latitudes above 60°
have to be covered 24 h a day, three Molniya satellites or two Tundra satellites are required. An
interesting variant is the three-apogee orbit with a 16 h period and an apogee at 43 500 km.
Table 2.6 presents the main features of Molniya and Tundra orbits. In the apogee position, which
is useful for frequent sampling, the height of the satellite exceeds GEO height.
12 GUIDE TO INSTRUMENTS AND METHODS OF OBSERVATION - VOLUME IV
Sats for
Coverage
Orbit type Inclination Period Apogee Perigee hemispheric
(from one satellite)
coverage
Molniya 63.4° 12 h ~39 800 km ~1 000 km Visible over 2 positions 3
for ~8 h
Satellites are injected into orbit by a launcher, which has to perform the following functions:
(a) To host the satellite in the fairing, where vital functions for the satellite are ensured. When
in the fairing, the satellite is stowed in a compact configuration to minimize volume
occupancy and to be protected against the effects of acceleration.
(b) To bring the satellite to orbit. In order to minimize the total mass brought to high altitudes,
the launcher is generally structured by stages. The first stage, which is the heaviest since it
has to provide the maximum thrust for lift-off, is separated early. The fairing is released at an
appropriate altitude. A further one or two stages are fired and separated in sequence.
(c) To release the satellite. For satellites in LEO circular orbits, the launcher releases the satellite
directly on the final orbit. For elliptical orbits, the launcher releases the satellite at the
perigee and provides it with a last acceleration to acquire the energy corresponding to the
intended orbit.
When in orbit, the satellite deploys its solar panels and starts autonomous operations. One
of the operations is to reach final orbit by activating its propulsion system. In the case of a
geostationary orbit, the satellite is released at a perigee in an elliptical orbit whose apogee is
35 786 km, and is equipped with an apogee boost motor. The motor, which uses solid, hybrid
or more often, liquid propellant (a liquid apogee motor), is fired at the apogee to provide the
acceleration necessary to circularize the orbit (see Figure 2.9).
Apogee
Low Earth
orbit
Transfer
Earth orbit
Perigee
Earth observation from space is mainly performed by exploiting electromagnetic radiation. The
few exceptions are in situ measurements at platform level (of gravity field, magnetic field, electric
field and charged particle density in the solar wind). This section deals with remote-sensing of
the Earth and focuses on:
(a) The electromagnetic spectrum and the ranges used for remote-sensing;
(b) The basic laws of interaction between electromagnetic radiation and matter;
2.2.1 The electromagnetic spectrum and the ranges used for remote-sensing
The spectrum of electromagnetic radiation as observed from space (with nadir viewing) is shown
in Figure 2.10. The displayed range (0.2 μm to 3 cm (or 10 GHz)) includes all that is used for
remote-sensing from space. The variation in wavelength is due to the interposed atmosphere,
with transmissivity ranging from 1 (atmospheric window) to 0 (full opacity due to total
atmospheric absorption).
Transmittance
Wavelength
Wave number
Spectrum subdivision Wavelength λ Frequency ν = c/λ
ν* = 1 /λ
UV Ultraviolet 0.01–0.38 μm 26 320–1 000 000 cm–1
VNIR Visible and near infrared (VIS + NIR) 0.38–1.3 μm 7 690–26 320 cm–1
Sub-
Submillimetre wave (part of FIR) 0.1–1 mm 10–100 cm–1 300–3 000 GHz
mm
Table 2.7 presents definitions of the subdivisions of the spectrum that are generally accepted,
though not standardized. In addition to the commonly used wavelength λ and frequency ν, the
wave number ν*, mostly used in spectroscopy, is also quoted.
A finer subdivision of the MW range and nearby FIR, used for radar but also by extension, for
passive radiometry, is provided in Table 2.8.
The overall spectrum shown in Figure 2.10 comprises five distinct regions, each with rather
different features.
Table 2.8. Bands used in radar technology (according to the American Society for
Photogrammetry and Remote-sensing)
0.6
0.4
0.2
0
0.4 0.5 0.6 0.7 0.8 0.9 1 1.5 2 2.5 3 3.5 4
Wavelength (µm)
Figure 2.11. Atmospheric spectrum in the range 0.4 to 4.0 μm. It includes several windows
and absorption bands from carbon monoxide (CO, about 2.3 μm), carbon dioxide (CO2, about
1.6, 2.1 and 2.8 μm), methane (CH4, about 2.3 and 3.4 μm), several oxygen bands (O2, mainly
about 0.77 μm), some nitrogen (N2) and ozone (O3) bands, and many important bands of
water vapour (H2O, mainly 0.94, 1.13, 1.37, 1.8 and 2.7 μm). Also shown is the molecular
continuum, which prevents using UV for Earth surface and low atmosphere
sensing from space.
In the UV region, atmospheric absorption is strong, mainly due to the major air constituents
(nitrogen (N2) and oxygen (O2)) and trace gases (the most important being ozone (O3)). The
Earth’s surface cannot be observed in that spectral region. The radiation source for remote-
sensing consists of reflected solar radiation.
The VIS, NIR and SWIR regions, from 0.4 to 3 μm and, in some cases, up to 4 μm, can be sensed
by means of reflected solar radiation. This range includes several transparent regions (windows)
and many absorption bands (see Figure 2.11).
In the MWIR and TIR regions, from 4 to 15 μm, the radiation source consists of the thermal
emission from the Earth’s surface and the atmosphere, which is mainly driven by water vapour
and carbon dioxide absorption/emission. These are important contributors to the greenhouse
effect. This thermal emission combined with the main atmospheric window enables the planet’s
thermal equilibrium to be maintained at agreeable values (see Figure 2.12).
Wavelength (µm)
16.67 12.50 10 8.33 7.14 6.25 5.56 5 4.55 4.17 3.85 3.57 3.33
300 CO 300
N2O
CH4
Brightness temperature (K)
280 280
H2O H2O
N2O
260 CO2 260
240 O3 240
220 220
CO2
600 800 1 000 1 200 1 400 1 600 1 800 2 000 2 200 2 400 2 600 2 800 3 000
Wave number (cm–1)
Figure 2.12. Atmospheric spectrum in the range 3.33 to 16.67 μm. The main atmospheric
windows are in the ranges 3.7 to 4.0 μm and 10 to 12 μm. There are large absorption bands
from water vapour (H2O) and carbon dioxide (CO2). Other species are: ozone (O3, about
9.7 μm), methane (CH4, about 7.7 μm), carbon monoxide (CO, about 4.6 μm) and
nitrous oxide (N2O, about 4.5 and 7.7 μm).
16 GUIDE TO INSTRUMENTS AND METHODS OF OBSERVATION - VOLUME IV
(A) Summer low latitude (D) Annual midlatitude (dry) with ozone
(B) Annual midlatitude (E) Annual midlatitude: nitrogen only
(C) Winter high latitude (F) Annual midlatitude: oxygen only
106
W W
W
104 W
W
W W
W
W W
O W
Zenith opacity (dB)
W W (A) W
W
O (B)
102
(C)
(D)
W
100
O
O O
O O
O (F)
(E)
10–2
O : Oxygen resonance
W : Water vapour resonance
(2) 100 200 300 400 500 600 700 800 900 1 000
Frequency (GHz)
Figure 2.13. Atmospheric spectrum in the range 2 to 1 000 GHz. The dominant feature is the
water vapour continuum that grows in opacity as frequency increases into the FIR range. This
prevents viewing the Earth’s surface at high frequencies from space. Invasive lines of ozone
also dominate the spectrum. The most useful features are the oxygen bands (about 57, 118,
388, 424, 487 GHz and above), used to infer atmospheric temperature, and the water vapour
bands (about 22, 183, 325, 380, 448 GHz and above).
Source: Klein and Gasiewski (2000).
The next spectral region, the Far IR, ranges from 15 μm to 1 mm (or 300 GHz). It is fully opaque
because of the water vapour continuum. In that region, which is difficult to explore because of
the lack of efficient detection techniques, there are absorption lines of several important spe-
cies, such as hydroxil radical (OH), known as a “cleaner” of the atmosphere, and hydrogen
chloride (HCl), a reservoir species that releases ozone-aggressive chlorine. Hydroxil radical and
hydrogen chloride are only observable in the FIR (for example, about 120 μm ≈ 2 500 GHz and
480 μm ≈ 625 GHz, respectively).
In the MW range, from 1 to 300 GHz, the radiation source consists of the thermal emission from
the Earth’s surface and the atmosphere. The atmospheric spectrum, starting from 2 GHz and
extending to submillimetre frequencies up to 1 000 GHz, is shown in Figure 2.13.
In the portion of the MW range where the atmosphere is more transparent, (that is, at
frequencies below about 100 GHz), the wavelengths exceed 3 mm and so are much larger than
cloud drop size, except for precipitating clouds. Therefore, the MW range is used for observing
the Earth’s surface or atmospheric properties in nearly all weather conditions.
Table 2.9. Spectrum utilization from active instruments (radar and lidar)
and radio occultation
Cloud and Rain radar, cloud Ku band (~13.6 GHz) and/or K a band (~35.5 GHz)
precipitation radar or W band (~94 GHz)
Aerosol, cloud Backscatter lidar UV-lidar (355 nm), VNIR-lidar (532 + 1 064 nm)
top
In this equation, ρ(λ,T,ζ,φ) denotes reflectivity that is the ratio of the backscattered radiation
I(λ,T,ζ,φ) to the incident radiation I(λ); τ(λ,T,ζ,φ) denotes transmissivity, or the fraction of I(λ)
that crosses the body; and ε(λ,T,ζ,φ) is the fraction of I(λ) that is absorbed by the body – it is
called emissivity for reasons explained below. The three coefficients depend on the radiation
wavelength λ, the body temperature T, and the observing geometry ζ, φ (zenith and azimuth
angles, respectively).
A body that is not reflecting and is totally opaque to radiation on any wavelength (where
ρ = τ = 0, and thus ε = 1) is called a black body. It radiates at any temperature, T and over the full λ
or ν spectrum, according to Planck’s law:
2π hc 2 1 2πν 3 h
B (λ , T ) = or B(ν , T ) = (2.7)
hc hν
λ 5
c 2
e λ kT −1 e kT −1
where:
h = 6.6256 · 10 –34 J s (which is the Planck constant);
c = 2.99793 · 10 8 m s–1 (which is the speed of light in a vacuum); and
k = 1.38044 · 10 –23 J K–1 (which is the Boltzmann constant).
B(λ,T) (or B(ν,T)) is the radiative power per unit surface over the hemisphere per unit of
wavelength (or frequency). The power radiated per unit of solid angle is B/π. The Planck function
for temperatures of 6 000 K and 273.16 K, which are representative of the Sun and the Earth’s
surfaces respectively, is shown in Figure 2.14.
18 GUIDE TO INSTRUMENTS AND METHODS OF OBSERVATION - VOLUME IV
X
10
8 V
UV
Figure 2.14. The Planck function for T = 6 000 K, representative of the Sun (higher curve), and
T = 273.16 K, representative of the Earth (lower curve); X values in ordinate are to
the power of 10.
The two curves in Figure 2.14 illustrate that the Sun’s and Earth’s surfaces have very different
radiative powers. However, after scaling down the Sun curve by the square of the distance
between the Sun and Earth, the two integrated areas become comparable, which reflects the
Earth’s radiative balance. If the upper curve in the figure is scaled, it is easy to see that the solar
radiation on Earth is very small for λ > 4 μm, while Earth radiation is negligible for λ < 3 μm. There
are significant amounts of solar and Earth radiation in the narrow interval between 3 and 4 μm.
A major difference between the two curves in Figure 2.14, after scaling, is the wavelength λmax
where the maximum emission occurs. This is given by:
b
λmax = with b = 0.0028981 m K Wien law (2.9)
T
Because of the double-logarithmic scale in Figure 2.14, it is difficult to appreciate how sharp
the Planck function is around λmax. In the case of solar radiation (T = 6 000 K) the peak emission
occurs around λmax = 0.5 μm and most of that power lies in the 0.2–3.0 μm range. In the case of
terrestrial radiation (T = 273.16 K, that is, T = 0°C) the peak is around λmax = 10 μm and most of
the power lies in the 3–50 μm range.
Another interesting feature of the Planck function is that, when moving from λmax to shorter
wavelengths, radiative power dramatically decreases, whereas when moving to longer
wavelengths, the decrease is more gradual (approximately two thirds of power occurs at
wavelengths longer than λmax). For very long wavelengths, or low frequencies, such as those in
the MW range, where the argument of the exponential in the Planck function is hν/kT << 1, the
term ehν/kT becomes ≈ 1 + hν/kT and the Planck function thus reduces to:
2π k ν 2
T B(ν , T ) =
Rayleigh-Jeans law (2.10)
c2
Because of this relationship, radiation measurements in the MW range can be considered
as temperature measurements and can be expressed as a brightness temperature (T B) in
temperature units. While the total radiative power changes with temperature in proportion to T 4
according to the Stefan-Boltzmann law (equation 2.8), that power changes in a linear way in the
MW portion of the spectrum. Conversely, when moving towards shorter wavelengths, the Planck
CHAPTER 2. PRINCIPLES OF EARTH OBSERVATION FROM SPACE 19
For example:
(a) For a body at 220 K (such as a cloud top in the upper troposphere), the radiation at 11 μm is
~1 200 times greater than at 3.7 μm; but at 300 K (surface), it is only ~130 times;
(b) The sensitivity to temperature changes, (that is, (∂B/∂T)/B) is three times higher at 3.7 μm
than at 11 μm.
As a consequence, 3.7 μm is well suited for surface observations, is optimal for fire detection, and
is less useful for high clouds.
The relationship represented by equations 2.7–2.10 is only valid for a black body (ε = 1).
For a common body, the relationship can be established through the following conceptual
experiment. If a number of bodies in an isolated system only exchange radiation among
themselves, it could be assumed that, after a transient time, each of them would reach
thermodynamic equilibrium, where the radiation each absorbs according to its absorption
coefficient ε(λ,T) is equal to the power it radiates P(λ,T). That is, P(λ,T)/ε(λ,T) = constant.
Assuming that one of the bodies is a perfect black body, this yields:
P (λ , T ) = ε (λ , T ) ⋅ B (λ , T ) Kirchhoff principle (2.11)
This shows that the absorption coefficient introduced in equation 2.5 also controls the body’s
emission properties, hence the name emissivity. Equation 2.11 indicates two important
consequences:
(a) At any wavelength and temperature, a body cannot radiate more than a black body at the
same wavelength and temperature;
(b) A body can radiate only at wavelengths at which it can also absorb.
Emissivity ε is a function of wavelength and, to a lesser extent, temperature. For certain bodies,
ε may be constant over large portions of the spectrum. If it is constant over the whole spectrum,
the body is called grey. The shape of the radiated power P(λ,T) is then exactly like B(λ,T),
although it is damped by a factor ε. The Wien law (equation 2.9) applies unchanged. The Stefan-
Boltzmann law becomes W(T) = ε · σ · T 4.
The Kirchhoff principle also applies to gaseous materials. Therefore, spectral lines of atmospheric
gases are generally (but not always) relevant to both absorption and emission.
Atmospheric windows are spectral regions where the atmosphere is nearly transparent. There is
no region where the atmosphere is fully transparent. All regions have some residual disturbance
from species that have a continuum, the most common of which is water vapour in the IR, MW
and, to some extent, the SW ranges. Another factor, particularly in SW, is scattering from dry air
molecules (mostly N2 and O2) and aerosols. Ultimately, the most transparent windows are:
τ ( λ, T ) · I (λ) Ɛ ( λ, T ) · B ( λ, T ) ρ ( λ, T ) · I (λ)
τ ρ
I (λ, ϴ)
ρ
I (λ)
τ
Equation 2.5 indicates that the coefficients ρ (reflectivity), τ (transmissivity) and ε (emissivity)
not only depend on radiation wavelength λ and body temperature T, but also on the geometric
condition (zenith angle ζ and azimuth angle φ) of the satellite platform with respect to the body.
In order to simplify the discussion, the following conditions are assumed: vertical viewing from
the satellite (Figure 2.15), flat surfaces and radiation towards the zenith (irradiance).
One component, τ(λ,T) · Iτ(λ), is transmitted radiation through the body. It can be found where
the body is not totally opaque and where there is a radiation source below it in the opposite
hemisphere compared to the satellite.
The component ε(λ,T) · B(λ,T) is emitted radiation, expressed through the Kirchhoff principle
(equation 2.11). It is always present unless the body is at absolute thermal zero. It is not present in
spectral regions, where the body does not absorb.
Taking account of all components, the radiation reaching the instrument on the satellite can be
expressed as:
I ( λ , T ) = τ ( λ , T ) ⋅ Iτ ( λ ) + ε ( λ , T ) ⋅ B ( λ , T ) + ρ ( λ , T ) ⋅ I ρ ( λ , θ ) (2.12)
There is considerable variation along the spectrum, so that the significance of the measurement
is reasonably stable only within narrow bandwidths around a specific wavelength (channels).
However, some general information is also contained in a wider range of wavelengths
(VIS + NIR + SWIR; MWIR + TIR; MW).
In this range, there is no thermal emission from the Earth (B = 0). Focusing first on the Earth’s
surface (land and ocean), the transmitted radiation is zero, since there is no source below
the Earth’s surface. Also when considering that reflectivity is nearly independent of body
temperature, equation 2.12 reduces to:
Isw ( λ ) = ρ ( λ ) ⋅ S ( λ ) ⋅ cos ζ (2.13)
CHAPTER 2. PRINCIPLES OF EARTH OBSERVATION FROM SPACE 21
Since the solar spectrum and observing geometry are known, the information carried by a
measurement in a SW channel is uniquely associated to surface reflectivity. Many geophysical
variables (vegetation parameters, ocean colour, texture of land) may be estimated by measuring
reflectivity at several wavelengths. However, clouds are the objects that are most in evidence in
SW. Equation 2.13 is not strictly correct for a cloud surface, since radiation from the underlying
surface can transmit through them. However, this effect has a limited impact given:
(a) The total transmissivity through (downward and upward) the atmosphere;
(c) The originating radiation source (the Sun) is stronger than any surface below the cloud;
(d) The reflectivity of the underlying surface is low (except for sand desert, snow and ice);
(e) The reflectivity of clouds ρ is generally higher than any other terrestrial surface.
Using SW-reflected radiation for quantitative purposes is not an easy exercise, since reflectivity
is generally anisotropic. The simplest case occurs when the body, whatever the direction of the
incoming radiation, and for any azimuth, homogeneously redistributes the reflected radiation
from the zenith according to a cosine-law. This is called Lambertian reflection. Fortunately, most
Earth surfaces observed from space at relatively large scale appear rather flat and rough, so that
Lambertian diffusion may be a good approximation. In many cases, however, the body exhibits a
bidirectional reflectance distribution function that should be measured a priori by observations
under different viewing directions and for different directions of the incoming radiation. The final
computation of the irradiance towards space requires hemispheric integration.
Using SW for Earth observation in atmospheric windows requires spectral sampling at several
wavelengths (channels) since at any one wavelength, several bodies may have signatures, and
one body may have a signature at several wavelengths. A multichannel capability is therefore
necessary to distinguish and simultaneously retrieve the different properties of different bodies.
For instance, clouds and snow have identical reflectance in VIS at 0.65 μm, but very different
reflectance at 1.6 μm. In addition, channel bandwidths must be appropriate to their purpose. The
most stringent are for ocean colour (Δλ ≈ 10 nm), and then vegetation (Δλ ≈ 20 nm), whereas for
other surface features and for clouds, bandwidths of several tens of nanometres are sufficient.
Another feature that affects the quantitative use of VIS + NIR + SWIR is polarization: specular
reflection tends to be privileged, with damping of the vertical component of the electric field.
The Stokes vector, which in the SW range consists of three components (polarization in three
directions with phase differences of 120°), fully describes the electric field, and so provides
important information about the body’s properties. Multipolarization is important for the
observation of those bodies that do not have strong multispectral signatures. Typical examples
would be aerosols and cirrus clouds (elongated ice crystals).
In the 4–15 μm range, solar radiation is virtually zero. On the Earth's surface (land and ocean),
transmitted radiation is also zero. Furthermore, considering that emissivity is nearly independent
from body temperature, equation 2.12 reduces to:
I IR ( λ , T ) = ε ( λ ) ⋅ B ( λ , T ) (2.14)
Equation 2.14 is also approximately valid for clouds, since the transmissivity of clouds in IR is
rather low (with the exception of thin cirrus).
22 GUIDE TO INSTRUMENTS AND METHODS OF OBSERVATION - VOLUME IV
The emissivity of most land surfaces, and certainly the ocean, is close to 1, with small variance.
Therefore the information acquired by a measurement in an IR channel is closely associated with
the Planck function (equation 2.7) or for a given wavelength, with the body’s temperature.
For a large variety of bodies where ε is close to 1, radiance observation in a TIR atmospheric
window enables rather accurate temperature measurement. It is particularly accurate in the
case of the sea, whose emissivity is close to 0.98. However, emissivity is not the only effect that
needs to be corrected. As previously mentioned, the atmospheric windows are not perfectly
transparent. For instance, the main window in TIR, 10–12 μm, is contaminated by the water
vapour continuum, particularly on the long-wave side. One way to reduce this disturbance is
to split the window into two channels, generally 10.3–11.3 μm and 11.5–12.5 μm. Differential
absorption is then used to estimate a correction (total column water vapour can also be
estimated as a by-product).
It is important to note that the 3.7 μm window behaves very differently in daylight and at night.
In daylight, it is strongly contaminated by reflected solar radiation, which needs to be subtracted
before using the channel for quantitative thermal emission estimates. As previously noted, the
3.7 μm window is much more sensitive to high temperatures than the 11 μm window. However,
the 3.7 μm window is of little use for low temperatures, such as those found in cloud tops in the
upper troposphere. The differential response to the temperatures of the 3.7 and 11 μm windows
can also be used for detecting fog at night.
As regards clouds, equation 2.14 is still approximately valid. However, with the exception of
very thick clouds (such as cumulus or nimbus-stratus), emissivity is substantially lower than
unity. Equivalent black-body temperature substantially underestimates true temperature,
and a correction must be applied to account for low emissivity. The usual method is to couple
the window channel with a channel that is strongly sensitive to water vapour. The difference
between the two T BB values indicates the cloud emissivity: the larger the difference, the lower the
emissivity.
The penetration of infrared radiation in clouds is very low. Measured temperature refers to the
top surface, and information about the interior of clouds is poor, especially for dense clouds.
However, the temperature of cloud tops that is in equilibrium with air temperature at the same
level is very important, because it indicates the altitude of the cloud in the troposphere, and
therefore the cloud type.
Information derived from black-body temperature about the cloud-top level is often inaccurate.
With thin clouds, such as thin cirrus, the background surface is much warmer than the air at
CHAPTER 2. PRINCIPLES OF EARTH OBSERVATION FROM SPACE 23
Bright V
i Stratocumulus Cirrus
s
i
b Arid soil Stratus
l
e Snow
Tilled soil
Cirrostratus
Cumulus
Sea
Infrared
Woody soil
Dark
Warm Cold
Figure 2.16. Scatterplot of VIS (0.65 μm) versus IR (11.5 μm), enabling the classification of
10 bodies. If projected onto one of either two axes, several clusters would be unresolved.
Source: Bizzarri and Tomassini (1976).
cloud level. As a result, surface radiation transmitted through the cloud adds to the cloud
emission, the cloud appears warmer, and the assigned level is underestimated. Conversely, thick
cirrus with high emissivity are observed as very cold, and may be confused with cumulonimbus.
In order to resolve such ambiguities, it is useful to plot VIS brightness and IR temperatures as
bi-dimensional histograms (Figure 2.16). By using only one band (the projection of the 2D
pattern on one axis), several clusters would be unresolved. By contrast, this example shows that
10 different objects can be identified through multi-band analysis. This is a simple example using
an old instrument (a very high resolution radiometer (VHRR)) and only two channels in VIS and
IR. Current multi-band analysis techniques can operate with many more channels.
In the MW range, solar radiation is virtually zero, but there is a diffuse source of incoming
radiation: the sky. Transmitted radiance is zero with regard to the Earth's surface, including land
and sea. There are only two contributions: thermal emission and reflected radiation. Those are
controlled by emissivity and reflectivity coefficients which, since τ = 0, are linked by the condition
ε + ρ = 1 (that is, ρ = 1 – ε). By expressing radiative power in units of temperature according to the
Rayleigh-Jeans equation (2.10), observed brightness temperature, T B, can be rendered:
Tsky(ν), sky brightness temperature, is composed of background cosmic radiation and the
contribution of precipitating clouds; it changes with frequency. In the main window and under
non-precipitating conditions, where ν ~ 40 GHz, Tsky(ν) may be ~140 K. Under heavy precipitation
conditions, Tsky may reach values as large as ~250 K.
24 GUIDE TO INSTRUMENTS AND METHODS OF OBSERVATION - VOLUME IV
The impact of Tsky(ν) strongly depends on the value of the emissivity, ε. Sea and land have
substantially different ε values.
The emissivity of the sea in the MW range is very low: ε ≈ 0.5. As a result, the two components in
equation 2.16 have equal weight. In the absence of precipitating clouds, Tsky is low and known:
the measurement can therefore be associated with sea-surface temperature. The optimal
frequency for sea-surface temperature is about 5 GHz (see Figure 2.17) where Tsky is much smaller
than that at 40 GHz. The measurement is rather accurate and is also applicable in all weather,
since the wavelength (λ = 6 cm) is much longer than any rain droplet. The 5 GHz sea-surface
signal intensity is representative of the temperature of a few millimetres of the deep-water
layer (sub-skin); that should be compared to a few tens of micrometres in the case of IR (skin
temperature). At higher frequencies, Tsky strongly increases, especially in the presence of heavy
precipitation. The high reflectivity value (1 – ε) is such that the observation is mainly an indicator
of precipitation.
Over land, emissivity is close to unity. The second component in equation 2.16 is therefore
ineffective, and precipitation is poorly detected. At higher frequencies (~90 GHz (λ = 3 mm)),
radiation welling up from the surface is scattered by large droplets, and even more so by ice
crystals in clouds. As a result, radiation reaching the satellite is decreased.
Polarization can also be used for measuring precipitation. Radiation reflected from the sea is
strongly polarized: when crossing a precipitating cloud, it undergoes depolarization that can
be measured to infer precipitation. The differential polarization can also be exploited over land,
since the emerging radiation scattered from droplets and ice crystals is polarized.
Observation under several polarizations may also be useful, regardless of the objective of
precipitation measurement. Differential polarization is sensitive to surface roughness – an effect
which must be taken into account when measuring sea-surface temperature. This can also be
used to infer wind speed over the sea, as indicated in Figure 2.17.
Figure 2.17 also indicates that MW radiation is sensitive to ocean salinity, but only at very low
frequencies, typically about 1.4 GHz (L band). The figure also shows that in order to measure
ocean salinity, it is necessary to account for sea-surface temperature and wind speed (or
roughness). Similarly to salinity, there is a water vapour absorption band that contaminates the
observation of temperature, wind and liquid cloud (precipitation). That band can also be used to
infer total column water vapour (precipitable water) over the sea. In summary, different variables
may have different signatures on various channels in the MW as well as the optical fields: multi-
channel analysis is therefore needed.
Liquid clouds
+
Water
Δ TB vapour
0
Δ Pi 10 20 30 40
Frequency GHz
– Sea-surface
temperature
Due to the totally different emissivity values of sea and land surfaces, the most obvious feature in
a MW image is the land/sea boundary. Since the emissivity of ice is close to unity, sea ice is also an
obvious observable in all weather conditions. MW images are particularly useful for geographical
regions that are often overcast. In cases where emissivity is close to unity over land, a decrease
in emissivity indicates the presence of water on the surface. This is because the emissivity of a
body is controlled by its dielectric constant: water on land is a salt solution, which increases
conductivity and thus decreases emissivity. This effect can be exploited to measure surface soil
moisture and snow properties.
Soil moisture measurements can be rather accurate on bare soil, but can decrease in accuracy as
vegetation increases. In order to penetrate vegetation, and to measure soil moisture at root level,
very low frequencies must be used, either in L or P band. At higher frequencies (above 10 GHz),
sensitivity to soil moisture is only significant if disturbance by vegetation is accounted for.
Two properties of snow are detectable in the MW range: surface melting conditions and, in the
case of shallow snowpack, water equivalent. In the latter case, relatively high frequencies are
preferred, as snow tends to be transparent at low frequencies. However, saturation can occur
at very high frequency signals in the upper layers of the snowpack. Several frequencies with
different penetration depths are therefore needed.
In an atmospheric absorption band, each layer of thickness dz absorbs radiation coming from
below and re-emits it. Assuming zero reflectivity of the atmosphere in IR, the atmospheric
transmittance from a height z to a satellite altitude H is given by:
H
− ε ( λ , z )⋅ N ( z )⋅dz
τ (λ , z ) = e ∫ z (2.17)
The radiative contribution of an atmospheric layer of thickness dz, at height z and associated with
a transmittance change of dτ (I,z), is: dI(λ,z) = B[λ,T(z)] · dτ(λ,z).
The radiation from the total atmospheric column to the satellite is:
1
I (λ ) = ∫τ (λ , z ) B[λ , T ( z )] ⋅ dτ (λ , z) (2.18)
s
Figure 2.18 shows that the transmittance (equation 2.17) tends to 1 as height z increases (that
is, as the thickness of the atmospheric layer between height z and satellite altitude H decreases).
This is accompanied by a decrease in emissivity ε and a decrease in concentration N of the
absorbing gas. The weighting functions have peak values that correspond to the inflection point
of the transmittance function. A simple way to read equation 2.20 is that each atmospheric
layer of thickness dz contributes to the radiation reaching the satellite according both to its
temperature (through the Planck function), and also to its effectiveness to contribute, as
quantified by the weighting function. The weighting function depends on the concentration
of the absorbing gas and the strength of absorption/emission (ε). The shape is such that low
26 GUIDE TO INSTRUMENTS AND METHODS OF OBSERVATION - VOLUME IV
Km
Km 48
43
32
39
27 36
32
7
23
27
20 6
7 6
5 23
16 20
4
16 5
13 3
2 4
13
7 3
1 7 2
4
4 1
0 0
0 0.2 0.4 0.6 0.8 1
(λ, Z ) κ ( λ, Z )
atmospheric layers are penalized because of absorption by the upper layers, and high layers are
penalized because of their low concentration of the absorbing gas. The atmospheric layer that
exhibits the greatest change in transmittance is usually the layer that contributes the most.
The inversion of equation 2.20 is not a trivial matter. It is a Fredholm equation of the second kind,
for which the existence or the uniqueness of the solution are not mathematically guaranteed.
In the present case, the existence is granted by nature. To ensure uniqueness of the solution it is
necessary to add constraints to it, since the problem is ill-conditioned. Many methods have been
developed since profile sounding from space began. Some are statistical and linear, others are
physical and non-linear, while others are a combination of both.
The next step is to retrieve the water vapour profile. Once CO2 absorption band channels have
been used to retrieve the temperature profile, H2O absorption band channels are used. The
main band is centred around 6.3 μm, and responds well to high temperature (in the low- to
mid- troposphere). For climate monitoring, it is important to measure water vapour in the
upper troposphere. But that requires using an 18 μm band, which is technologically difficult to
construct due to a lack of efficient detectors in the FIR range.
It is not easy to retrieve a water vapour profile or, more generally, to retrieve the concentration
profile of an absorbing gas. The weighting functions of the absorbing gas peak at varying
altitudes in the atmosphere, depending on concentration and remote-sensing frequency.
CHAPTER 2. PRINCIPLES OF EARTH OBSERVATION FROM SPACE 27
Further difficulties arise where clouds are present. If the instantaneous field of view (IFOV)
is entirely filled by a cloud with uniform features, a profile may still be retrieved by the same
method, although it will only cover the atmosphere above the cloud. If only a fraction η of the
IFOV is filled by a cloud of emissivity εcloud, the transfer equation becomes:
I (λ ) = (1 − ε cloud ⋅η ) I clear (λ ) + ε cloud ⋅η ·I cloud (λ ) (2.21)
where εcloud · η is the effective cover.
Several methods deal with the effects of clouds. One starts the retrieval process by using the
channels with weighting functions that peak above the cloud-top level to achieve a first-guess
profile. The first guess is then iterated by changing effective cover values, until the measurements
in all other channels fit best. Another method compares a number of nearby IFOVs on the
assumption that the signals differ only because of the different fractional covers η, and then
extrapolates to zero η.
In any event, it is acknowledged that, when cloud cover in the IFOV exceeds approximately 20%,
no attempt should be made to retrieve profiles in IR. The IFOV of sounding instruments used to
be several tens of kilometres. Fortunately, that has now been reduced to the order of 10 km, so
that the probability of finding a substantial number of cloud-free IFOVs in a given area is much
higher.
The problem of clouds is greatly alleviated in the MW range, where sounding is possible for all
weather conditions except heavy rain. The species of well-known and constant concentration
used for temperature profile retrieval is O2, with absorption bands in the 50–70 GHz range and
at about 118 GHz (not yet used from a satellite). For water vapour, the 183 GHz band can be used
effectively. The 22 GHz band provides a weak signal that can provide total column-integrated
amount over the sea. There are other absorption bands for temperature and water vapour at
higher frequencies, but the radiative effect of the water vapour continuum makes it impossible to
observe the troposphere using those spectral bands.
The transfer equation in the MW range is essentially a simpler version of the IR equation (2.20):
instead of the Planck function (equation 2.7), it is possible to use the Rayleigh-Jeans
approximation with linear temperature dependence (equation 2.10).
The question may arise as to why the MW band is not exclusively used for temperature and
humidity sounding, as it performs in nearly all weather conditions. This is because vertical
resolution requires a high sensitivity to temperature variations (with height). Vertical resolution is
best in the 4.3 μm band, where the Planck function varies roughly in relation to T 12. In the 15 μm
band, there is less sensitivity because the Planck function varies in relation to T 5. In the MW
range, since B is a linear function of T (see equation 2.10), sensitivity (∂B/∂T)/B varies in relation
to T –1 (it decreases as temperature increases). One interesting feature of the various bands is
that, whereas the 4.3 μm band is well suited to the lower troposphere, and the 15 μm band is
well suited to the middle and high troposphere, the MW band at 57 GHz is better suited to the
stratosphere.
Current instruments for the MW range have already reached maximum vertical resolution
performance: it cannot improve beyond ~1.5 km in the mid-troposphere, and will be worse in the
lower troposphere because of strong contamination from the ground.
In the short-wave (SW) range (UV, VIS, NIR, SWIR) absorption band measurements are mostly
used for atmospheric chemistry through spectroscopic methods. The radiative transfer equation
is more complicated than equation 2.20. Instead of thermal radiation as described by Planck's
law, the more complex process of scattering is used. Retrieving geophysical variables relies on
modelling rather than explicit equations. As well as being used in atmospheric chemistry, the
analysis of absorption bands is used for other purposes (see the spectrum in Figure 2.11), such as
deriving:
(a) Atmospheric pressure at the Earth’s surface: this is derived from estimates of the total
column of oxygen in the band around 0.77 μm compared with nearby windows. It is one
of the very few approaches available to measure surface pressure from space. Accuracy is
limited by the scattering effect of aerosols, implying that the measurement also provides
information on aerosols.
(b) Cloud-top height: this is derived from a deficit that arises when the total column of
oxygen is measured; the deficit itself is the result of cloud masking the lower part of the
column. In principle, this is more accurate than calculating the cloud-top height from the
equivalent black-body temperature in IR, correcting for cloud emissivity, and transforming
temperature into height using a temperature profile.
(c) Lightning: a very narrow bandwidth channel at 0.774 μm is used. Strong absorption from
oxygen obscures the Earth's surface and enables flashes to be detected even in daylight.
The intensity and number of flashes in a given period over a given area are representative of
convection, and so serve as a proxy for precipitation. In addition, lightning activity causes
NOx to be generated in the atmosphere, and reflects the Earth’s electric field.
(d) Total column water vapour: the signal in one or more of the water vapour bands (about 0.94
or 1.37 μm) is compared with a signal from nearby windows. This can be more accurate
than using IR or MW profiling.
Figure 2.18 shows how weighting functions become broader as height increases. This indicates
that the vertical resolution of temperature and humidity profiles using passive IR or MW
radiometry degrades with increasing altitude. The resolution obtained from using spectrometers
is currently considered adequate (~1 km) in the mid-troposphere; but it becomes marginal
(~2 km) at tropopause level, where much better resolution is required. In the stratosphere, the
vertical resolution degrades further and rapidly becomes unusable. Two techniques offer help:
limb sounding (including through occultation of the Sun, moon or stars) and radio occultation.
In cross-nadir sensing mode, vertical resolution is determined by the sharpness of the weighting
functions, which is in turn controlled by spectral resolution. In limb mode, vertical resolution
is determined by mechanical scanning, that is, by the instrument IFOV across the atmosphere
when viewed transversally in the Earth’s limb region (Figure 2.19). Vertical resolution depends on
the step change rate, which is tuned to the instrument viewing aperture and to the intensity of
available radiation. It is generally set to between one and three kilometres. Horizontal resolution
is relatively inaccurate, since the measurement is integrated over a large optical path, as shown
in Figure 2.19. The total optical path may be thousands of kilometres long, but the effective path,
once weighted by atmospheric density, extends to some 300–500 km around the tangent point.
The sources of radiation are solar radiation reflected from the atmosphere, or atmospheric
thermal radiation in the IR or MW ranges. In general, limb observations address not only
temperature and humidity profiles, but also trace gases for atmospheric chemistry purposes.
CHAPTER 2. PRINCIPLES OF EARTH OBSERVATION FROM SPACE 29
Platform
Line of sight
1
..
.
15
16
17
Atmospheric
layers
Earth
In the SW range (UV, VIS, NIR, SWIR), the atmosphere can be scanned by directly pointing to
the Sun while the Sun is setting or rising (occultation). Observation is conducted by measuring
the damping of spectral lines in the solar spectrum. Sun occultation has the great advantage of
avoiding any mechanical movement of the instrument telescope, and any calibration. That is
because the spectra measured during occultation are compared to the solar spectrum measured
shortly before (or after) occultation under the same conditions. One disadvantage is that, at least
for polar-orbiting satellites, coverage is limited to high latitudes, where a satellite can observe
sunrise or sunset as it enters or leaves the night arc of its orbit. More extended coverage is
possible by using moon occultation, while all latitudes can be covered through occultation of the
stars. However, less radiation is available in those cases.
The sensing methods described above assume that the sources for remote-sensing are reflected
solar radiation and the Earth’s emitted thermal radiation (plus other minor sources, such as
background sky radiation in MW and the moon or stars in occultation). These natural sources
enable passive sensing, for which observation wavelengths are largely determined by natural
targets. In active sensing, the source is artificial and the sensing wavelength is not entirely driven
by the physical properties of the target. Instead, the wavelength can be chosen, while also
taking account of signal generation and propagation constraints. The following active sensing
principles are considered:
(a) Radio occultation (for high vertical resolution profiles of temperature and humidity);
(b) Radar (for altimetry, scatterometry, cloud and precipitation, and imagery);
(c) Lidar (for clouds and aerosols, air motion, altimetry, and atmospheric chemistry).
The change of propagation direction due to refraction by the crossed atmosphere (bending
angle α) is converted into a phase shift. The shift is accurately measured and then converted into
a refractivity profile during the occultation process, which lasts approximately 90 s.
30 GUIDE TO INSTRUMENTS AND METHODS OF OBSERVATION - VOLUME IV
LEO
𝛼
GPS
a r
a LEO
Earth
600–1 000 km
a = an impact parameter
r = distance from centre of curvature to tangent point
Figure 2.20. Principle of radio-occultation sounding. During the setting or rising of the GPS
satellite over the horizon of the LEO satellite, the refraction induced by the crossed
atmosphere changes the direction of propagation by a bending angle α.
The phase shift is ultimately a time measurement, one of the most accurate measurements in
physics. The other measurement is the distance between satellites. Since time and distance are
fundamental metric quantities, radio occultation provides absolute measurements (not requiring
calibration): this is a very attractive feature in terms of climate monitoring. In fact, long-term
observations of radio occultation are considered a benchmark among methods of detecting
climate change.
Radio occultation data are difficult to process for two reasons: first, the position of the tangent
point (see Figure 2.20, right hand panel) moves during the profile measurements; and second,
pressure, temperature and humidity are not measured independently. Therefore, 4D assimilation
into a numerical weather prediction model is needed. It is less complex to retrieve temperatures
from the upper troposphere and stratosphere, since water vapour content is very low.
Temperature retrieval is similarly straightforward in the lower troposphere, where water vapour
is responsible for most variance.
The vertical resolution of radio occultation profiles in the upper troposphere and stratosphere,
(approximately 0.5–1.0 km) cannot be matched by cross-nadir viewing IR or MW measurements
(1.5–2 km). In addition, the frequencies used (L band: see Table 2.9) are insensitive to clouds,
even if precipitating. As a result, although essentially performed in limb mode, the measurement
can be extended down to the Earth's surface, in order to observe, for example, atmospheric
discontinuities such as the top of the planetary boundary layer. Furthermore, radio occultation is
one of the few methods that can infer surface pressure: this is done by correlating the height of
the tropopause and the air pressure at ground level.
In order to account for the signal rotation induced by the ionosphere, transmissions from
navigation satellites exploit at least two nearby frequencies. As by-products of the correction
process, information relevant to space weather is obtained, such as total electron content and
electron density profile.
CHAPTER 2. PRINCIPLES OF EARTH OBSERVATION FROM SPACE 31
2.2.5.2 Radar
Radars (radio detection and ranging) transmit pulsed signals to the object to be observed,
and collect the backscattered signal. In essence, radars measure distance (or range) and the
backscattered power resulting from the radar reflectivity or radar cross-section of the body.
The radar equation may be written in different forms. The easiest to understand is:
Pt ⋅ G 1
Ps = ⋅σ ⋅ ⋅ Aeff (2.23)
4π ⋅ r 2
4π ⋅ r 2
where Ps is the power backscattered to the antenna, P t is the power transmitted by the antenna,
G is the antenna gain, Aeff is the effective area of the radar receiving antenna, r is the distance, and
σ is the radar cross-section. Therefore:
The antenna gain may be expressed as G = 4π · Aeff/λ2 (a relation that derives directly from the
diffraction law). Expressing the effective area of the radar’s receiving antenna Aeff from this
alternative expression and inserting it into equation 2.23 yields:
Pt ⋅ G 2 ⋅ λ 2
Ps = ⋅ σ (2.24)
64π 3 ⋅ r 4
Different types of radar favour the measurement of either range accuracy (altimeters) or
reflectivity/cross-section accuracy (scatterometers). Radar for clouds and precipitation focus on
both range (for vertical profiling) and reflectivity. One feature that can be emphasized is image
resolution by synthetic aperture radar.
Radar altimetry
The main purpose of altimetry is to measure the sea level and to map it so as to determine ocean
dynamic topography. The radar characteristics are optimized to enhance range measurement
as much as possible. Sea level is measured in terms of the time taken for a radar pulse to reach
the sea surface and return to the satellite. Because sea level is computed as distance from the
satellite, the satellite needs to be located with extreme accuracy. One or more of the following
systems is used to ensure precise orbits:
(a) Laser tracking of the satellite by ground stations and laser-reflecting mirrors on the satellite;
One drawback of radar altimetry is that viewing must be limited to nadir-only; otherwise echoes
from surrounding areas interfere with time analysis. As a result, the observing cycle is very long.
Corrections to the signal are requested to account for: ionospheric rotation (the two frequencies
used are: ~13.6 GHz (main) and ~5.3 GHz (support)); and water vapour (a co-aligned MW
radiometer is used at ~23 GHz (main) and ~35 GHz and/or ~19 GHz (support)).
In addition to measuring the range, an altimeter also records and analyses fluctuations and
measures the intensity of the echo. The observations provided are:
(a) Significant wave height: derived from analysis of the spread in time of the collected echoes;
(b) Sea level: derived from filtering wave-related fluctuations and considering the
instantaneous satellite altitude with respect to the geoid;
32 GUIDE TO INSTRUMENTS AND METHODS OF OBSERVATION - VOLUME IV
(c) Wind speed: derived from analysis of the fluctuation of the intensity of the echoes;
(d) Improved knowledge of the geoid: derived from long-term statistics of observed sea levels;
(e) Total electron content: derived as a by-product of the correction for ionospheric rotation.
Radar scatterometry
Unlike radar altimetry, where the focus is on range measurement, radar scatterometry optimizes
the accuracy of the measured radar cross-section σ (see equation 2.24), which is often normalized
and called σ 0 (sigma-naught). While the ranging subsystem may be absent from the instrument,
calibration must be extremely accurate.
The radar cross-section is a function of the target’s dielectric property, the viewing geometry and
the incident radiation (wavelength, polarization). Scatterometers are mainly used to derive sea-
surface wind. The target is capillary waves, which are closely associated with wind stress. Sigma-
naught changes with wind speed, relative wind direction and sight line. By measuring σ 0 under
several azimuth angles, both speed and direction can be determined.
The relationship between σ 0 and wind is complicated: the practical solution is empirical or semi-
empirical. Furthermore, it is not a unique relationship in terms of direction: with two viewing
angles, several ambiguities remain (fewer ambiguities remain with three angles). When σ 0 values
are directly assimilated into a numerical weather prediction model that accounts for wave–
atmosphere interaction, ambiguities are solved by the model.
The differences between wind measurements taken with scatterometers and those taken with
passive MW radiometers can be summarized as follows: (i) passive MW generally provides
information on wind speed only; information on direction can only be acquired if several
radiometric channels are equipped with full polarization capability; (ii) information from
scatterometry is generally better quality, especially for low wind speeds (less than ~3 m/s);
however, for high speeds (more than ~20 m/s) passive MW may perform better.
Primarily designed for sea-surface wind, scatterometers provide several kinds of observation:
(a) Sea-surface wind in all weather conditions (C band) or nearly all weather conditions
(Ku band);
(b) Air pressure over the sea surface (achieved by applying geostrophic relations to wind
maps);
(c) Soil moisture in scarcely vegetated areas (using C band, and occasionally Ku band);
While the radar altimeter focuses on ranging and the radar scatterometer focuses on the radar
cross-section, cloud and precipitation radar emphasizes both. Ranging is necessary to measure
the vertical profile of cloud particles, while σ is required to infer the concentration and size of
reflecting particles. However, the accuracy required for ranging in order to obtain a precipitation
profile is of the order of 100 m, instead of 1 cm for altimetry.
CHAPTER 2. PRINCIPLES OF EARTH OBSERVATION FROM SPACE 33
For rain droplets, provided that their diameter D is less than λ/10 (that is, in Rayleigh-scattering
conditions), the radar cross-section is:
π5
σ= ⋅ K 2 ⋅ D6 where K is the dielectric constant (2.25)
λ4
The total backscattered radiation received by the radar is the sum of all reflectors of all diameters
in the IFOV. Assuming a Marshall-Palmer distribution (N0 · e–ΛD) for the particle diameters, total
reflectivity can be expressed as:
Dmax
π5
Z = ⋅ ∫ K 2 ⋅ N 0 ⋅ e− Λ D ⋅ D 6 ⋅ dD (valid for D λ ) (2.26)
λ 4
0
Equations 2.25 and 2.26 are generally applicable to ground-based meteorological radar that use
S band (~10 cm) or C band (~5 cm). They are to be compared with D ~ 0.5 cm, which is typical of
precipitating clouds. Those frequencies are not used from space, as the corresponding IFOV on
the ground that has a reasonably-sized antenna is too coarse. Due to frequency regulations, the
frequencies that can be exploited by a spaceborne radar are ~14 GHz (~2 cm), ~35 GHz (~0.9 cm)
and ~94 GHz (~0.3 cm). Therefore, equations 2.25 and 2.26 are not fully applicable, but must be
corrected in a complex way to account for Mie scattering conditions.
Once the reflectivity Z is measured, there are several ways to convert Z into precipitation rate R.
First, it is necessary to infer the precipitation rate at the surface. That cannot be directly measured
from space, but must instead be derived from measured properties in the vertical column
associated with the precipitation profile.
Cloud and precipitation radar is the only technique that can provide measurements of the cloud
base, an important variable for aeronautical meteorology and for climate. The accuracy and
reliability of the measurement depends on the radar frequency; the radar must also penetrate the
full cloud thickness. From an operational viewpoint, cloud and precipitation radars have several
disadvantages, particularly their limited swath, which prevents frequent observing cycles. And
so, although passive MW imagery continues to be used as the basis for frequent precipitation
observation, the accuracy of precipitation data from passive MW radiometry still needs to be
improved. The continuing availability of at least one radar in space is necessary for “calibrating”
the system of passive MW radiometers, along the lines of the concept for the Global Precipitation
Measurement mission.
In the MW range, spatial resolution is limited by diffraction. For a side-looking radar that views
θ° off nadir, at orbital height H, with an antenna diameter L, and assuming a flat surface, the
IFOV is:
H ⋅c
IFOV = 1.24 ⋅ (2.27)
cos θ ⋅ L ⋅ν
Table 2.10 shows the IFOV and L relationship for radar in several bands, assuming that
θ = 23° and H = 700 km (parameters of the SAR on SeaSat, Figure 2.21). It illustrates that the
requirements for IFOV = 1 km would be very difficult to meet, and requirements for IFOV = 100 m
or below would be impossible with a real-aperture antenna concept.
In the SAR concept (Figure 2.21) the antenna is elongated along the satellite motion. Its narrow
dimensions determine the swath and are parallel to the sub-satellite track. The longer side
determines the area where signals are to be analysed. The radar footprint corresponds to that
of a real-aperture antenna, while the situation is varied for the resolution elements in the field
(pixels). The pixels across the swath are at different distances from the satellite. The satellite
can locate them due to its ability to distinguish small-range changes. The echoes from the
pixels ahead of the sub-satellite cross-track line are affected by a positive Doppler shift (where
a frequency is higher than the one transmitted); the echoes from the pixels behind that line
undergo a negative Doppler shift. By capturing the moment of shift inversion, the pixels can be
assigned to a location along the satellite track.
34 GUIDE TO INSTRUMENTS AND METHODS OF OBSERVATION - VOLUME IV
SAR may be used in different operating modes, depending on the trade-off between resolution
and swath, and on the combination of transmitted and received polarizations. One operating
mode is designed for wave spectra. A small part of the image (a vignette) is sampled at intervals.
With Envisat, for example, the vignette is 5 km x 5 km at the best spatial resolution of 30 m, and
sampled at 100 km intervals along the track. The echoes in the vignette are analysed to plot the
power spectra, from which it is possible to determine the dominant wave direction, the dominant
wave length/period and power associated with significant wave height.
The list of SAR applications is long, although not all bands are suitable for all applications:
(e) Snow melting conditions and snow water equivalent (preferably X band);
(f) Surface soil moisture (preferably L band, especially for the areas around plant roots);
(g) Vegetation type (preferably C band) and total biomass (preferably P band);
Other important applications are possible through interferometry because the phase control of
the SAR signal is extremely accurate. Signals from different orbits of the same satellite or different
satellites can be accurately co-registered in order to implement interferometry. This enables, for
instance, land surface topography to be obtained for an improved digital elevation model, and
iceberg heights to be measured.
By using interferometry between passes of the same satellite in an orbit with a repeat cycle, it
is possible to measure changes such as iceberg drift, variations of glacier cover and lake extent,
volcanic surface topography changes and bradyseisms, coastal erosion and urbanization.
Satellite track
7 450 m/sec
DAT DAT
DR δ AT ≥
2
6° cone
(1° cross-cone) Satellite ground track,
C
along track (AT) δR≥
SAR antenna or azimuth direction 2ß
800 km Rf
23°
Rn
Resolution element
Nadir
δR δ AT
Grazing
angle ψ
Cross track
or range direction (R)
Radar footprint
100 km Rf – Rn 𝜆R
ground swath (Wg) Wg = =
cos ψ DR sin ψ
2.2.5.3 Lidar
While the principle of lidar (light detection and ranging) is the same as that of radar, the
electromagnetic range is different: MW is used for radar, while SW is used for lidar. Most lidars
make use of wavelengths in the UV (for example, 355 nm), VIS (for example, 532 nm), NIR
(for example, 1 064 nm) or SWIR (for example, 1 550 nm) ranges; longer wavelengths (for
example, 10.6 μm) may also be used. The source is a laser (light amplification by stimulated
emission of radiation). It is extremely directional, but the large distance between the satellite and
the Earth means that significant electric power is needed and that large telescopes are required
to collect the backscattered signal. The use of lidar in space therefore presupposes considerable
resources. Although in principle, lidar could be used to scan an area for imaging purposes, lidar
systems in space have so far only been nadir-pointing or monodirectional. The following missions
are lidar-based:
Backscatter lidar
Backscatter lidar primarily deals with the observation of aerosols. This implies the use of
wavelengths that are similar in size to very small aerosols (~1 μm). In order to capture more
36 GUIDE TO INSTRUMENTS AND METHODS OF OBSERVATION - VOLUME IV
aerosol properties (size, phase, absorption/scattering ratio, and ultimately, type), more concepts
have been developed. Some are based on the use of two wavelengths, while other are based on
high spectral resolution laser in order to distinguish Mie and Rayleigh scattering components. In
any case, a backscatter lidar is a very large instrument (the telescope may have ~1 m aperture).
The footprint may be as small as a few tens of metres, but more decorrelated echoes are included
to increase signal-to-noise ratio. Such echoes will help to ensure that the final resolution is in the
range of a few hundred metres. The vertical resolution determined by the lidar ranging system is
set to a few hundred metres.
Designed primarily for aerosols (the most demanding mission), the backscatter lidar enables
different kinds of observations to be made, including:
(a) Aerosol profile and aerosol properties (size, phase, absorption/scattering ratio, type);
(b) Cloud-top height (much more accurately than with passive NIR and IR techniques);
(c) Optical thickness of thin clouds (cirrus) and cloud base of semi-transparent clouds;
(e) Atmospheric discontinuities such as tropopause and top of the planetary boundary layer, as
revealed by the change of refractivity index.
Doppler lidar
Doppler lidar deals primarily with the observation of wind profiles, a key variable for weather
forecasting.
The current operational technique available for observing wind profile consists of tracking the
movement of clouds or water vapour patches in frequent images, either from geostationary
satellites, or from polar orbiting satellites in polar areas with frequent satellite overpasses.
This limits the opportunity to take measurements only at altitudes where tracers are present
(generally one layer, sometimes two). Hyperspectral IR sounding in GEO is expected soon: when
it becomes available, frequent water vapour profiles will be available, and water vapour patterns
will be tracked at several heights. However, vertical resolution and accuracy are expected to be
limited, and coverage will not include high latitudes.
Cloud-motion tracking is the only technique available for cloudy areas, while the Doppler lidar
enables tracking to be conducted in clear air. The tracer consists of eddies of the turbulent
atmosphere, of aerosols and of molecular scattering.
Exploitation of the Doppler shift due to the wind implies oblique viewing. The laser pulse
repetition frequency is such that the corresponding IFOV may be less than 100 m. However, as
with any radar or lidar, a number of decorrelated echoes need to be averaged to improve signal-
to-noise ratio: the final resolution is an IFOV as large as several tens of kilometres. Because of the
limited availability of electric power, the instrument has a limited duty cycle; for instance it may
sample at 200 km intervals along a line parallel to the sub-satellite track. The vertical resolution
of the wind data collected depends on the time sampling rate of the return echo. This is adjusted
for a vertical resolution of about 1 km in the mid-troposphere, less than 1 km in the planetary
boundary layer, and more than 1 km in the stratosphere. No technique based on cloud and water
vapour tracking from GEO can compete with this performance. The anticipated accuracy of the
wind horizontal component is less than 2 m/s.
A Doppler lidar is a very large instrument (the telescope may have a 1.5 m aperture). It requires a
dedicated satellite in an orbit lower than that usually used for meteorological satellites.
For wind lidar, coverage is a major limitation. Since one instrument only covers a line parallel
to the sub-satellite track, a constellation of satellites would be required to provide frequent
CHAPTER 2. PRINCIPLES OF EARTH OBSERVATION FROM SPACE 37
coverage. As long as the sustainability of such a constellation cannot be achieved, the baseline
system for wind profile will continue to be imagery and hyperspectral sounding from GEO, with
Doppler lidar providing support for calibration of the overall system.
Another difficulty with wind lidar is that measurement can only be along-sight (1D). That means
that, in order to retrieve the required 2D horizontal component, assimilation into a numerical
weather prediction model is necessary.
The Doppler lidar is primarily designed for wind, which is the most demanding mission. It
provides several kinds of observations, including wind profile in clear air or in the presence of
thin cirrus, aerosol profile (from echo intensity) and cloud-top height.
Lidar altimeter
Radar altimeters can provide measurements as accurate as a few centimetres. However, at above
20 km, their horizontal resolution is quite coarse. By implementing SAR processing of the along-
track signal, the resolution can be brought to ~300 m, which is still insufficient for accurately
detecting boundaries. Another limitation of radar altimeters is their unsuitability for observing
surfaces with high emissivity (and thus low reflectivity) in the MW range, such as land and ice.
Lidar does not have those limitations. The horizontal resolution can be a few tens of metres
and the vertical resolution less than 10 cm. That fine resolution makes it possible to capture the
border between sea water and polar ice and (after successive passes) to profile the height along-
track in order to map ice thickness.
In order to operate over the sea, even though reflectance in the VNIR range is very low, a lidar
altimeter must have a large telescope (aperture ~1 m). For improved cost-effectiveness, a sensor
using a second wavelength is usually added to observe the atmosphere (for example, 1 064 nm is
used for surface and 532 nm is used for atmosphere). In this case, a lidar altimeter operates as an
ordinary backscatter lidar.
As with any other altimeter, the lidar altimeter requires extremely accurate orbit determination,
since the basic ranging measurement provides the distance of the object from the satellite in
orbit. Precise orbit determination is achieved by a GPS receiver, and laser ranging from a network
of ground stations with an array of retroreflectors on the satellite.
(f) Cloud-top height, optical thickness of thin clouds and cloud base of semi-transparent
clouds;
(h) Atmospheric discontinuities such as tropopause and top of the planetary boundary layer.
38 GUIDE TO INSTRUMENTS AND METHODS OF OBSERVATION - VOLUME IV
The conditions for DIAL to be applicable are: that the species is relatively abundant; that the
absorption line is strong and lies in a spectroscopically “clean” region; and that there is a nearby
“clean” atmospheric window. Examples of possible applications are:
(a) CO2 by exploiting lines at 1.57 μm in the 1.6–1.7 μm window or 2.05 μm in the 2.0–2.3 μm
window;
Missions exploiting DIAL are only at the stage of proposal or feasibility study. Since the spectral
bandwidth (and thus the energy available for the observation) is very narrow and the reflector (a
gas) is very weak, the instrument has to be very large and overall measurement is technologically
challenging.
Earth observation from space implies a complex system composed of (i) a space segment to
perform observations, and (ii) a ground segment to manage the space segment and process
observation data.
(c) The communication tools to receive commands and convey the instrument output to the
ground.
The size and/or mass of satellites for Earth observation can range over two orders of magnitude:
(a) Nanosat: < 10 kg (actually unlikely to be used for operational Earth observation);
The satellite platform hosts the instruments and provides several services:
(a) Power supply for instruments, telecommunications, and all other satellite subsystems;
(d) Thermal control to keep the instruments within specified operating conditions;
(e) Housekeeping devices to monitor and control the status of all satellite subsystems;
(h) Processing capability to handle instrument data and format data streams to be transmitted;
(k) Communication facilities to transmit observational and housekeeping data to the ground;
(l) Other communication services, where the platform has a data relay function only.
Navigation and positioning systems are necessary for geolocation of observed data, both during
viewing, and afterwards for ground processing. The following systems are used:
Laser retroreflectors
These are mirrors which tend to be corner cubes. They reflect laser beams sent to the satellite
by laser-equipped ground sites during positioning sessions. Laser retroreflectors are used on
many satellites for a posteriori precise orbit determination. This is achieved by post-processing
a number of measurements in night-time and clear sky only. The analysis involves a full network
of coordinated ground stations. The results are sparse and only available after a certain delay.
However, they are so accurate that they can be used for space geodesy applications.
Radio-positioning systems
These systems are specifically designed to support altimetry missions. They comprise radio links
between the satellite on the one hand, and ground transmitting and/or receiving stations on
the other. Positioning is performed in near real time and, with improved accuracy, after post-
processing. Two examples of such systems are:
40 GUIDE TO INSTRUMENTS AND METHODS OF OBSERVATION - VOLUME IV
(b) Precise Range and Range-rate Equipment (PRARE), which measures differential signals from
a network of ground stations.
GNSS receivers
These systems make use of the phase difference of signals from several satellites in the GNSS.
The GNSS includes the navigation satellite constellations of the United States of America (GPS),
the Russian Federation (GLONASS), the European Union (Galileo), and China (Compass, known
as Beidou in Chinese). A large number of satellites currently use GNSS receivers to support their
navigation. Positioning is performed in real time.
Star trackers
These are charge-coupled device imagers that track bright stars, recognize their pattern,
and send information to a satellite’s attitude control system. Star trackers provide continuous
monitoring of satellite attitude much more accurately than systems based on horizon-sensing.
This is necessary for instruments that require accurate pointing information (such as limb
sounders), both for active attitude control during flight and for subsequent instrument data
processing. An increasing number of satellites are now being equipped with star trackers.
The orientation and stabilization systems are primary platform features that determine
instrument pointing capability.
The side of the platform on which the sensors are placed should ideally be kept facing the Earth’s
surface, unless the satellite mission has a different purpose (such as monitoring the Sun). Since
the platform tends to keep a steady orientation in relation to the stars during its orbital motion, a
stabilization mechanism is required.
Three-axis stabilization is much more suitable for maintaining a constant orientation towards
the Earth and also supports more instruments on one platform. This allows for active control of
the satellite attitude with respect to rotations around: the axis perpendicular to the orbital plane
(pitch), the axis tangent to instantaneous motion in orbit (roll), and the nadir direction (yaw).
Active control is critical, because it implies accurate attitude determination (by, for
example, horizon detection, star trackers or GNSS receivers) and efficient actuators (such
as micropropulsion devices, gyros, very fine-angle change detectors and efficient control
electronics). A loss of active control is among the primary causes of mission failure. Active control
may affect data quality because of limited accuracy, particularly in the case of high-resolution
instruments and high orbits (GEO), and mechanical perturbation of instrument pointing
associated with turning on the actuators.
CHAPTER 2. PRINCIPLES OF EARTH OBSERVATION FROM SPACE 41
In addition to the main orientation and stabilization systems (spinning and three-axis control),
smart attitude control systems are also in use, especially with small satellites. For instance, the
gravity-gradient system uses a long boom that tends to the nadir direction, and thus keeps one
side of the platform pointing towards the Earth’s surface.
A basic trade-off for satellite design lies between the capabilities that are implemented on
board and those that can be achieved on the ground if sufficient information regarding on-
board features is available. Hardware implementation on board may be expensive, prone to
irrecoverable failure, and provide limited performance. Therefore, it is advisable to reserve
hardware implementation to cases where it is absolutely indispensable and reasonably safe.
Moreover, the housekeeping system provides all the ancillary information necessary to accurately
process data on the ground.
The housekeeping system manages both the platform (deformations, temperature of radiant
surfaces, attitude, status of power generators and all other subsystems) and the instruments
(status and temperatures of the various parts, control signals for electronics, etc.). In general,
instrument housekeeping is at least partially implemented inside the instrument itself.
The amount and completeness of housekeeping constitutes a discriminating factor for the
class of a satellite. Operational satellites are equipped with plenty of housekeeping devices for
subsystem monitoring and the possible activation of recovery manoeuvres, such as by switching
to redundant units. Housekeeping information is also a basic element of accurate instrument
calibration and data georeferencing.
Notwithstanding the importance of a good housekeeping system, there are limitations to the
accuracy of what can be achieved by such software processing. The residual errors of software
corrections or reconstructions may exceed what is allowed by the application. Therefore, certain
corrections need to rely on on-board hardware.
The platform must transmit to the ground the observation results from the various instruments.
Whatever the satellite height over the Earth’s surface, radio transmission to the Earth will have to
cross the ionosphere and plasmasphere, which block the propagation of electromagnetic waves
with frequencies lower than the critical plasma frequency (~25 MHz). Direct visibility is needed
between the transmitters and the receivers both on the satellite and the ground station.
The simplest method of collecting observed data from a satellite is by direct broadcast in real
time. For a LEO satellite, a ground station will acquire all the data that a satellite transmits when
passing within the acquisition range. The size of the acquisition range is the same as a satellite’s
FOV for zenith angle ζ = 90° (in principle), or ζ = 85° (with a reduced risk of interferences from
ground sources or occlusion from orography). Table 2.1 shows that, for a satellite height of
800 km for instance, the acquisition range is a circle with a diameter of 5 000 km (for ζ = 85° or
elevation = 5°).
Where the satellite velocity is in the region of 400 km/min and the satellite pass is centred over
the acquisition station, the acquisition session lasts 15 min at most, and is reduced to a few
minutes for peripheral passes.
This type of acquisition is the most convenient because it provides the observed data to the user
in real time for immediate processing. However, only data observed and transmitted during the
satellite pass inside the acquisition range can be acquired by the local receiving station. For GEO,
direct broadcast can be continuously received by a station located within the FOV.
An alternative way for a LEO satellite to receive data is to store the observed data on board and
transmit them on command, when the satellite overflies a central acquisition station. The central
42 GUIDE TO INSTRUMENTS AND METHODS OF OBSERVATION - VOLUME IV
station, also used to send commands to the satellite, has the same acquisition range as any local
station; if it is placed at a high latitude, it can collect data from many orbits. Table 2.4 shows that
most Sun-synchronous orbits pass less than 10° from the pole, so that a central station placed
at a latitude of, for instance, 80° will intercept all orbits and provide global data acquisition. The
store-and-dump acquisition method has the advantage of enabling data recovery for the whole
globe, but also has several drawbacks:
(a) Access to data is slower, since the delay includes the time needed to: run the whole orbit
(up to 100 min), receive at the central station (about 10 min), relay to the central processing
facility (about 10 min) and redistribute to the users (about 10 min). The total delay in the
availability of data is therefore 2–3 h.
(b) The satellite has to transmit the data accumulated in one orbit (about 100 min) during the
time it spends within the acquisition range of the central station (about 10 min). Therefore
data rate and bandwidth need to be one order of magnitude higher than for direct read-
out, which heavily impacts on the cost, size and complexity of the station. This acquisition
mode is suitable for a satellite operator but generally not suitable for an individual user.
(c) There are instruments with data rates so high that they cannot be fully stored on board:
selection of data to be stored may be needed, either by reducing the resolution or by prior
selection of frames (for example, the local area coverage (LAC) mode of NOAA Polar-
orbiting Operational Environmental Satellites (POES)).
If not acquired locally in real time, the data gathered in a central processing facility need to
be retransmitted to the users, generally after preprocessing. In the case of GEO satellites, a
transponder on the same satellite can be used for data retransmission to local user stations.
For LEO satellites, between the two extreme cases (direct read-out providing data over a limited
area in real time, or store-and-dump of global data with 2–3 hours’ delay), there are alternative or
complementary data recovery schemes, which can use:
(a) Several downlink stations spread over the globe, including, for instance, one near each
polar region; this reduces the length of time that data need to be stored on board the
spacecraft;
(b) A network of direct read-out stations spread over the globe; each one acquires data on
limited areas and retransmits them to data centres; this reduces the delay in the availability
of data to a few tens of minutes, but does not necessarily achieve global coverage; or
(c) A data relay satellite that receives data in real time from observing platforms and relays
them to a central processing facility; this reduces the delay in the availability of data to a few
minutes.
Data recovery timeliness is a critical issue for operational satellites; that is particularly the case
for meteorology, because of the coexisting requirements for timeliness and global coverage.
For research and development applications, the timeliness requirement is less stringent, and
the store-and-dump method tends to be used in conjunction with an effective archiving and
retrieving facility, which provides advanced data stewardship.
In addition to providing Earth observation data from a platform in orbit, satellites can support
other services, and so act as a telecommunications relay. The most common forms of such relays
are:
(a) Data collection from in situ platforms located on the ground, on aircraft, on balloons,
on buoys, on ships and even on migrating animals. The data collection platform (DCP)
may transmit all the time, at fixed intervals, or upon interrogation from the satellite.
CHAPTER 2. PRINCIPLES OF EARTH OBSERVATION FROM SPACE 43
Mobile platforms may be located from the satellite if in LEO. GEO satellites serve either
DCPs within their line of sight (regional) or DCPs carried on mobile platforms (ships,
aircraft, etc.) that migrate among the views of different GEO satellites (international).
(b) Search and rescue distress signals are detected from small pieces of transmitting equipment
carried by those in distress. The request for help is then relayed to one of the centres
of a worldwide search and rescue network. Search and rescue missions are conducted
cooperatively by several operational meteorological satellites in LEO and GEO. The service
from LEO is called the Search and Rescue Satellite-aided Tracking System (SARSAT), while
the GEO-based service is known as Geostationary Search and Rescue (GEOSAR).
(a) The central station for satellite command and global data acquisition;
2.3.2.1 Central station for satellite command and global data acquisition
This element of the ground segment may be generically termed the command and data
acquisition station (CDA). Typical tasks of a CDA are:
(a) To collect command sequences from the mission and operation control centre and uplink
commands to the satellite (for payload configuration, satellite configuration, orbit control,
etc.);
(b) To acquire satellite telemetry data (for attitude and orbit determination, satellite and
payload status, etc.) and immediately deliver it to the mission and operation control
centre(s);
(c) To acquire geophysical and ancillary data (housekeeping, calibration, etc.) and deliver it to
the data processing and archiving centres;
(d) To index the acquired data streams with accurate time and orbital elements.
It is possible to have only one CDA for geostationary satellites. For near-polar satellite systems,
it is possible to avoid blind orbits by placing the CDA at a very high latitude (such as Svalbard at
78°N). In order to improve the timeliness of the availability of observed data, auxiliary stations
may be used (such as an Antarctic station). For low-inclination orbits, a network of CDAs is
necessary, with one acting as the main CDA.
CDAs use S-band frequencies (about 2 100 MHz) to command the satellite. S band is nearly
insensitive to weather and less critical to pointing accuracy. For geophysical data acquisition,
44 GUIDE TO INSTRUMENTS AND METHODS OF OBSERVATION - VOLUME IV
the L band is used (about 1 700 MHz) if the data rate is below 10 Mbps; otherwise either the
X band is used (about 8 GHz) for data rates of up to some 100 Mbps; or the Ka band (about
26 GHz) is used for data rates of several hundreds of Mbps.
These elements may be generically termed the operation control centre (OCC). Their tasks are:
(a) To collect information on satellite, payload, and orbit status from the CDA and (in the case
of the orbit only) from other ranging stations;
(b) To collect requirements for elements including payload configuration and measurement
sequence planning from data processing and archiving centres, and from other users
entitled to input requirements into the mission plan;
(c) To analyse the information on satellite, payload, and orbit status, and on payload or mission
configuration requirements; to generate instrument performance monitoring reports; to
elaborate the operations plan and deliver commands to the CDA for uplinking the satellite;
(d) To provide data processing and archiving centres with ancillary information, which is
relevant for data processing and results from activities relating to operations, payload or
mission control (including accurate orbit determination, satellite attitude behaviour and
payload status).
Mission control centres are closely connected with users, application centres and scientific teams.
OCCs are closely connected with CDAs and the units responsible for satellite development.
OCCs also have full knowledge of satellite design features. The two centres are often co-located,
although this is not compulsory, and should preferably be secured by a back-up centre.
(a) Acquiring geophysical, calibration and selected auxiliary data from CDAs;
(b) Acquiring auxiliary data on orbit, satellite and payload from OCCs;
(g) Analysing mission status, payload status and requirements for mission planning;
Core products are normally generated by the satellite operator at a central facility. External
specialized centres may supplement those by processing other specific products.
Satellite data archiving requires the maintenance of high levels of hardware availability for data
ingest and storage, as well as discovery and retrieval services, with provision for long-term
data preservation over decades. Data must be associated with metadata that contain all the
information necessary to use and evaluate the data. Comprehensive, standardized metadata, and
standardized globally interoperable catalogue systems enable data discovery to be extended to a
worldwide scale within the WMO Information System.
CHAPTER 2. PRINCIPLES OF EARTH OBSERVATION FROM SPACE 45
Depending on data volumes and timeliness requirements, several methods of accessing data and
products are available:
(a) Direct read-out from the satellite (when available; particularly prevalent among LEO
satellites). This provides the best timeliness but presupposes the capability to receive raw
data on an appropriate receiving station and to preprocess that data with an adequate
software package.
(c) Near-real-time retransmission of data via specialized networks such as the WMO Global
Telecommunication System.
(d) Active FTP retrieval from data centres for off-line data, specifically from archives.
Data and product distribution may be subject to conditions, depending on the status and data
policy of the space agency running the programme (operational, research and development,
commercial) and what the programme will be used for. Access to data retransmission services is
generally controlled by encryption and subject to registration, even if no charges are levied.
User stations are installed to make use of real-time or near-real-time data transmission from the
satellite. Depending on the satellite access modality and the user requirement, there may be:
(a) High-data-rate acquisition stations for full reception of the data available either by open
access or by agreement with the satellite owner;
(b) Low-data-rate acquisition stations for a data selection of either reduced volume or quality;
(c) Receiving terminals of commercial telecommunication satellites used for data dissemination
after preprocessing or processing in the data-processing and archiving centre.
Frequencies used by the high-data-rate acquisition stations are in the L band (about 1 700 MHz)
if the data rate is below approximately 5 Mbps; otherwise the X band (about 8 GHz) is used
for data rates of up to approximately 100 Mbps. The low-data-rate acquisition stations make
use of relatively low frequencies (L band: about 1 700 MHz for GEO, and very high frequency
(VHF): about 137 MHz for LEO) that can be used on mobile stations, such as ships. Commercial
telecommunication satellite terminals use Ku band (about 11 GHz) or C band (about 3.8 GHz).
46 GUIDE TO INSTRUMENTS AND METHODS OF OBSERVATION - VOLUME IV
Satellite observations are retrieved from the raw data acquired from the instruments though a
processing chain. Different processing levels are usually referred to, the detailed definitions of
which depend on the instrument in question. Table 2.11 provides a generic description of these
processing levels.
Table 2.11. Generic description of processing levels (to be adapted to each instrument)
Level Description
Instrument and auxiliary data reconstructed from satellite raw data after removing
0
communications artefacts.
Instrument data 1a (for LEO) or 1.0 (for GEO): calibration and geolocation attached but
extracted, at full not applied.
1 original resolution, 1b (for LEO) or 1.5 (for GEO): calibration and geolocation applied.
with geolocation and
calibration information. 1c, 1d, etc: optional for specific instruments.
2 Geophysical product retrieved from a single instrument in the original projection.
Geophysical product retrieved from a single instrument, mapped on uniform space and time grid
3 scales, possibly on a multi-orbital (for LEO) or multi-temporal (for GEO) basis. Irreversible process
due to resampling.
4 Composite multi-sensor and/or multi-satellite products; or result of model analysis.
Level 0 data are processed from the raw data stream by removing communication artefacts (such
as synchronization frames and communication headers) and appending all necessary auxiliary
data, including housekeeping and station-added information on timing and tracking. Level 0
data should be archived permanently to enable reprocessing with an updated instrument model
(such as improved calibration or georeferencing).
Level 1a or 1.0 data consist of instrument files (counts) in the original instrument projection,
with an appended (but not applied) deformation matrix or algorithm for georeferencing and
calibration coefficients. The process from Level 0 to Level 1a/1.0 is fully reversible. Level 1a/1.0
data are normally permanently archived, although in principle they could be reproduced if
Level 0 data have been archived.
Level 1b or 1.5 data consist of calibrated, co-registered and geolocated data in physical units
(generally radiances), still in the original instrument projection. The process from Level 1a/1.0
to Level 1b/1.5 is not reversible because of truncation, discretization and resampling operations.
Although Level 1b/1.5 may in any case be reprocessed from Level 1a/1.0 or Level 0, the
processing effort is such that, in general, Level 1b/1.5 data are permanently archived.
Level 1c data are processed from the Level 1b data of certain instruments to enable end users
to make use of that data. The process may be fully reversible (for example, spectra from
interferograms by Fourier transform) but equally may not be (such as with apodized spectra). In
general, these data are permanently archived. For certain instruments, further Level 1 steps (1d,
1e, etc.) may be defined (the addition of a cloud flag, for instance).
Level 0 and Level 1 processing is performed by the satellite operator. Where there is a
direct readout, the satellite operator generally ensures the availability of Level 0 and Level 1
preprocessing software to local data users.
Level 2 products are generated from Level 1 data by applying algorithms that make limited use of
external information. Data quality information is appended. These products are generated in the
original instrument projection and tend to be permanently archived.
Level 3 products are generated by compositing a sequence of Level 2 products from successive
orbits (with LEO) or at successive times (with GEO). Possible gaps in the sequence may be filled
CHAPTER 2. PRINCIPLES OF EARTH OBSERVATION FROM SPACE 47
Level 4 products are generated by blending data from different instruments on the same or
different satellites, either with other data sources, or by assimilation in a model. In Level 4
products, the contribution of a specific satellite instrument may be hardly recognizable.
CHAPTER 3. REMOTE-SENSING INSTRUMENTS
This chapter provides an overview of basic instrument concepts. It introduces the high-level
technical features of representative types of Earth observation instruments.
A large variety of instruments and sensing principles are used in Earth observation. The key
instrument features introduced here are:
The most characteristic feature of an instrument is the way in which it scans the scene to acquire
the necessary observations. That depends on the type of orbit (low Earth orbit or geostationary
Earth orbit), on the platform attitude control (spinning, three-axis stabilized), and sometimes on
the specific type of measurement being taken. Only the most common scanning techniques will
be described in this chapter.
A driving requirement for scanning is whether the scene should be observed with continuity
(imagery) or can be sampled (sounding). A similar scanning mode may be used in both cases.
However, imagery requires continuous scene coverage, while sounding can accommodate
sampling with gaps.
In LEO spacecraft, a 2D Earth scene scanning can use satellite motion for an along-track
dimension. A cross-track scan can then be provided by a rotating scan mirror (Figure 3.1).
Rotation speed is synchronized with satellite motion so that the cross-track image lines come out
contiguously.
In the case of GEO, there is no satellite motion with respect to the Earth’s surface. The two
scanning dimensions must therefore be generated either by the instrument or by satellite
rotation. In Figure 3.2, it is assumed that the satellite is spin-stabilized: the west–east scan is
Chopper
In-flight calibration
(reference black body)
PbSe
45°
Filter
Cooling patch
Secondary Primary
mirror mirror
provided by the satellite rotation, while the north–south scan is obtained by a step motor. For a
three-axis stabilized platform, the instrument must generate both west–east and north–south
scans.
The advent of array detectors has led to additional scanning possibilities. With LEO, a linear array
can be placed orthogonally to the satellite track, and can scan the scene without any mechanical
movement (pushbroom scanning). Under another scheme, the linear array can be placed parallel
to the track; cross-track mechanical scanning will then scan several lines in parallel (whiskbroom
scanning). With GEO, whiskbroom scanning is now the rule. For instance, SEVIRI on Meteosat
Second Generation scans three lines in parallel for the IR channels, and nine lines for the high-
resolution VIS channel.
A disadvantage of conical scanning is that, with the selected incidence angle, the field of view
does not normally reach the horizon. For example, the swath from an 800 km orbital height
is ~1 600 km for a typical zenith angle of 53°, which is optimal for enhancing differential
polarization information in MW. By contrast, the swath for a cross-track scanning instrument
is close to 3 000 km, assuming a ± 70° zenith angle range, as shown in the present volume,
Chapter 2, Table 2.1.
The swath is an important feature of an instrument in LEO since it determines the observing
cycle. Chapter 2, Table 2.2 of the present volume outlines that, for a Sun-synchronous orbit at
800 km, one instrument with a swath of at least 2 800 km provides one global coverage per day
for measurements operated in daytime only (for example, SW sensors), or two global coverages
per day for measurements operated day and night (for example, IR or MW sensors). MW conical
scanning instruments generally provide one global coverage per day.
50 GUIDE TO INSTRUMENTS AND METHODS OF OBSERVATION - VOLUME IV
Satellite orbit
Fore view
Footprints
45°
Flight
Swath width direction
Fore view
The order of magnitude of the observing cycle (Δt, in days) for a given swath can be estimated
by a simple calculation. Considering the Equator’s length (~40 000 km), the number of orbits
per day (~14.2) and assuming that there is no significant overlap between adjacent swaths at the
Equator, the calculation is as follows:
(a) For day and night sensors (IR or MW) operating on both ascending and descending passes:
Δt = 1 400/swath (for example, for a MW conical scanner with 1 400 km swath: Δt = 1 day);
(b) For daytime-only sensors (SW) operating on only one pass per orbital period:
Δt = 2 800/swath (for example, for VIS land observation with 180 km swath: Δt = 16 days).
The concept of swath is not applicable for instruments with no cross-track scanning, such as
altimeters or cloud radars. Where that is the case, the cross-track sampling interval Δx at the
Equator replaces “swath” in the relationship above. It is also useful to estimate the global average
of this sampling interval. The interval is given by a slightly different relationship because of
shorter orbit spacing at higher latitudes:
(a) At an average cross-track sampling interval Δx, the typical time needed for global
coverage is:
(i) Δt = 900/Δx for day and night sensors (for example, the Environmental Satellite
(Envisat) Radar Altimeter – 2 (RA‑2): Δx = 26 km, Δt = 35 days);
(ii) Δt = 1 800/Δx for daytime only (for example, the NOAA Solar Backscatter Ultraviolet
(SBUV) instrument: Δx = 170 km, Δt = 11 days).
(b) Reciprocally, in a time interval Δt (for example, the orbit repeat cycle or a main sub-cycle)
the average cross-track sampling interval obtained is:
(i) Δx = 900/Δt for day and night sensors (for example, the JASON altimeter: Δt = 10 days,
Δx = 90 km);
(ii) Δx = 1 800/Δt for daytime only (for example, NOAA SBUV: Δt = 5 days, Δx = 360 km).
CHAPTER 3. REMOTE-SENSING INSTRUMENTS 51
Limb sounders are generally considered as non-scanning instruments in the cross-track direction.
Assuming a cross-track sampling interval Δx = 300 km which is equal to the horizontal resolution
of the measurements, and taking into account the Michelson Interferometer for Passive
Atmospheric Sounding (MIPAS) that flew on the Envisat satellite as an example, the relationships
above yield the following observing cycle:
Δt ≈ 3 days for day and night sensors (for example, MIPAS on Envisat).
For instruments providing sparse but well-distributed observations, the coverage cycle or the
average sampling can be estimated by comparing the number of events and their resolution with
the Earth’s surface to be covered. In the example of radio occultation using GPS and GLONASS,
each satellite is able to provide about 1 000 observations per day, with a typical measurement
resolution of 300 km for a total Earth’s surface of 510 million km2. Therefore:
(b) Number of satellites required for an observing cycle Δt: N = 5.7/Δt (for example, for
Δt = 0.5 days, the number of satellites is close to 12).
Sun-, moon- or star-occultation instruments are an extreme case. Sun or moon occultation
provides very few measurements per day, and only at the high altitudes of the day/night
terminator in the case of the Sun, or at somewhat lower latitudes for the moon. Star occultation
may provide several tens of measurements per day (for example, 40 for the Envisat Global Ozone
Monitoring by Occultation of Stars (GOMOS)), evenly distributed by latitude.
Another major characteristic feature of an instrument is the spectral range over which it operates.
As discussed in the present volume, Chapter 2, 2.2, the spectral range determines which of the
body’s properties can be observed including reflectance, temperature and dielectric properties.
Within the spectral range, there may be window regions and absorption bands, which mainly
address condensed or gaseous bodies, respectively.
A spectral range may be more or less narrow, depending on the effects to be enhanced, or on
the disrupting factors that need to be eliminated. The sub-divisions of a band or a window
covered by an instrument are called channels. The number of channels depends on the number
of pieces of independent information which have to be extracted from one band. If a limited
number of well-separated channels are sufficient for the purpose, the instrument may only
include those channels, and it is called a radiometer. If the information content rapidly changes
with the frequency along the spectral range to the extent that channels must be contiguous, the
instrument is called a spectrometer.
The technique adopted for channel separation, or for spectrometer subrange separation, is a
major characteristic of the instrument. There are essentially two possible ways to physically
separate two channels towards individual detectors (or detector arrays) and associated filter
systems. First, the beam could be focused on a field stop and split into two bands by a dichroic
mirror. The advantage is that co-registration is ensured, as the two channels look at the same
field of view (which could comprise an array of IFOVs). However, if the two wavelengths are too
close to each other (for example, a split window), a dichroic mirror cannot separate them sharply
enough. The second solution is to let the full beam produce the image in the focal plane and set
detectors (or detector arrays) with different filters (thus identifying different channels) in different
parts of the focal plane (in-field separation). It is a much simpler solution, but as each channel
looks at a different IFOV, co-registration problems can occur. Combined solutions are possible:
Figure 3.4 illustrates the solution implemented in the Meteosat Second Generation SEVIRI to
separate the eight IR channels. Each channel is viewed by three detectors. Current detector
arrays are much larger than they have been in the past, which makes in-field separation more
convenient.
52 GUIDE TO INSTRUMENTS AND METHODS OF OBSERVATION - VOLUME IV
7.3 µm
IFOV A
13.4 µm
9.7 µm
8.7 µm
IFOV C
12.0 µm
3.8 µm
IFOV D
10.8 µm
Spectrometers provide continuous spectral sampling within the spectral range or in a number
of spectral subranges. (These subranges are sometimes called channels and should not be
confused with the channels of a radiometer). There are several types of spectrometer, the
simplest of which is the prism. Others include grating spectrometers and interferometers; the
most common interferometers are the Michelson and the Fabry-Pérot. Figure 3.5 shows the
scheme of a Michelson interferometer.
Fixed mirror
Beam
Input port 1 splitter
SCENE
Moving
mirror
Output port 1
For example, in IASI, an instrument mounted on the Metop satellite, the excursion of the moving
mirror is ± 2 cm. Therefore, OPDmax = 4 cm and Δν = 0.25 cm–1. If fine analysis of the spectrum is
needed (for instance, to detect trace gas lines), apodization that implies a factor ~2 is required.
The apodized resolution is therefore Δν = 0.5 cm–1.
For a grating spectrometer, the resolution is determined by the number of grooves, N, and the
chosen order of diffraction, m. The resolving power λ/Δλ is given by:
λ ∆λ = m ⋅ N (3.2)
For a Michelson interferometer, the spectral resolution is constant with a variable wavelength.
For a grating spectrometer, however, it is the resolving power that is constant, and the spectral
resolution that changes with wavelength. If a grating spectrometer is to cover a wide spectral
range, that range should be subdivided into subranges that use different orders of diffraction, m.
Resolving power may thus change based on subrange.
For a radiometer, the number of channels and their bandwidths play an equivalent role to
spectral resolution for a spectrometer.
3.1.3 Spatial resolution (instantaneous field of view, pixel, and the modulation
transfer function)
The IFOV is probably the closest to what is commonly meant by resolution. In optical instruments
(that is, SW and IR instruments), it is determined by the beamwidth of the optics and the size of
the detector. In MW it is determined by the size of the antenna.
In optical systems, the size of the IFOV is designed primarily on the basis of energy considerations
(see 3.1.4). The IFOV may be determined by the shape of the detector; and although that may be
square, the contour of an IFOV is not unduly sharp. In fact, the image of a point is a diffraction
figure called point spread function (Figure 3.6). The IFOV is the convolution of the point spread
function and the spatial response of the detector.
The energy entering the detector is also determined by the integration time between successive
signal readings. During image scanning, the position of the line of sight will change by an
amount called sampling distance. When plotted in a rasterized 2D pattern, a pixel (picture
element) appears as a series of rectangular elements. In the x‑direction, they correspond to
the sampling distance and in the y‑direction, to the satellite motion during the time distance
from one line to the next (or the step motion in the north–south direction from GEO). The pixel
is often confused with the resolution. That is because users can directly perceive the size of
the picture element, whereas the IFOV is an engineering parameter that the user cannot see.
Nevertheless, it is wrong to think that resolution can be improved by reducing the integration
time, since the integration time must be suitable to ensure an appropriate radiometric accuracy
(see 3.1.4).
There is a balance between the size of the IFOV and the size of the pixel. When using a “perfect”
imager, sampling is performed so that the sampling distance is equal to the IFOV size. That
means that the IFOVs are continuous and contiguous across the image. Otherwise the image
may be oversampled (pixel < IFOV, that is, there is overlap between successive IFOVs) or
undersampled (pixel > IFOV, that is, there are gaps between successive IFOVs). Oversampling
is useful to reduce aliasing effects (undue enhancement of high spatial frequencies because
of reflection from the border). Undersampling may be necessary when more energy must
be collected in order to ensure the required radiometric accuracy. Examples of relationships
between IFOV and pixel are:
(a) Advanced Very High Resolution Radiometer (AVHRR) – IFOV: 1.1 km; pixel: 1.1 km along-
track, 0.80 km along-scan (oversampled);
(b) SEVIRI – IFOV: 4.8 km; pixel: 3.0 km across-scan and along-scan (oversampled).
MTF is closely linked to the concept of IFOV and pixel. It is even more directly linked to the size
“L” of an instrument’s primary optics. The MTF represents the capability of the instrument to
correctly manage the response to amplitude variation at the scene. It is the ratio between the
observed amplitude, and the true signal amplitude from the scene, figured as sinusoidal. The
observed amplitude is damped due to factors such as the diffraction from the optical aperture,
the “window” introduced by the detector and smearing from integrating electronics. The effect
of integrating the radiation over the (squared) detector introduces a contribution to the MTF:
( )
MTFwindow ( f ) = sinc(π ⋅ IFOV ⋅ f ), with f = 1 ( 2 ∆ x ) km −1 and sinc y =
sin y
y
(3.3)
This shows that, for Δx = IFOV, MTF = sinc (π/2) = 2/π. Therefore, even for a “perfect” imager,
MTF is lower than unity. The value 2/π ≈ 0.64 corresponds to the area of a half sinusoid inscribed
in a square. The concept of MTF must be seen as closely associated with radiometric accuracy: it
specifies at which spatial wavelength two features are actually resolved if their radiation differs
by just the detectable minimum. Two features whose radiation differs by substantially more than
the detectable minimum can be resolved, even if they are substantially smaller. However, if they
are as small as IFOV/2, then MTF = 0, and they can in no way be resolved (f = 1/IFOV is called the
cut-off frequency).
It is interesting to note how MTFwindow changes at different spatial wavelengths measured in terms
of IFOV (where spatial wavelength is 2 Δx). Table 3.1 can be derived from equation 3.3:
This shows that features as small as two thirds of the IFOV can be resolved, but only if their
radiances differ by more than three times the detectable minimum. It also shows that features
twice as large as the IFOV can be resolved if their radiance difference exceeds the minimum by
10%.The other major contribution to the MTF is diffraction. The relation is:
2 −1 f f f
MTFdiffraction =cos ( ) − 1 − ( )2 (3.4)
π fd fd f d
where f = 1/(2 ϑ H) (with H = satellite height); ϑ = angular resolution (that is, IFOV/H); fd = L/
(λ H); L = aperture of the primary optics; and thus f/fd = λ/(2 ϑ L).
CHAPTER 3. REMOTE-SENSING INSTRUMENTS 55
Diffraction is dominant when the wavelength λ is relatively large (as with microwaves), when the
optics aperture is relatively small, or when the satellite altitude is relatively large (as with GEO).
The value MTFdiffraction = 0.5 occurs for f/fd = 0.404, that is:
λ
ϑ = 1.24 (3.5)
L
which is the classical law of diffraction.
In summary, what is commonly meant by the term “resolution” involves at least three
parameters. Although they should be considered in context, each one is more closely associated
with a different perception:
(a) IFOV: not visible to the user; controls the radiometric budget of the image;
(b) Pixel: provides direct perception of the degree of detail in the image;
(c) MTF: by monitoring the amplitude restitution, provides the perception of contrast.
Although scarcely visible to the user, radiometric resolution is a defining element of instrument
design. Scanning mechanisms, spectral resolution, spatial resolution, integration time and optics
apertures are all designed to fulfil radiometric resolution requirements. Radiometric resolution
is the minimum radiance difference necessary to distinguish two objects in two adjacent IFOVs.
The observed difference is a combination of the true radiance difference between two bodies
(signal) and the difference observed even when the contents of the IFOVs are identical (noise).
The signal-to-noise ratio (SNR) is one way of expressing radiometric resolution.
Noise is a function of several factors, as set out in the radiometric performance formula:
2F
NESR = (3.6)
D * ⋅∆ν ⋅τ ⋅ π ⋅ t ⋅ ∆Ω
where:
NESR = noise equivalent spectral radiance (unit: W m–2 sr –1 (cm–1)–1, that is, per unit of wave
number);
F = f/L, F-number (f = system focal length, L = telescope aperture);
D* = detectivity (strongly depending on ν);
Δν = spectral resolution (expressed in terms of wave number ν = 1/λ);
τ = instrument transmissivity;
t = integration time;
ΔΩ = system throughput, given by the product of (π·L2/4) by (IFOV2/H2); where:
π·L2/4 = areal aperture of the telescope;
H = satellite altitude;
IFOV2/H2 = solid angle subtended by the IFOV.
In general, when defining I(ν) = spectral radiance at instrument input (unit: W m–2 sr –1 (cm–1)–1),
this leads to:
SNR = I (ν ) NESR (3.7)
For SW, the input radiance is the solar spectral radiance corrected for the incidence angle
and reflected according to body reflectivity (or albedo, if the body can be approximated as a
Lambertian diffuser). Equations 3.6 and 3.7 lead to:
SNR
∝ L (at a specific input radiance I (ν ) or albedo ρ ) (3.8)
IFOV ∆ν ⋅ t
That relationship explicitly links user-oriented parameters such as SNR, spectral resolution Δν,
IFOV and integration time t to the size of the primary optics L.
56 GUIDE TO INSTRUMENTS AND METHODS OF OBSERVATION - VOLUME IV
In the case of broadband channels, the concept of NESR, as expressed in equation 3.6, must be
redefined in terms of integrated noise over the full spectral range of each channel. In that case, it
is also possible to obtain a relationship similar to formula 3.10:
1
NE∆R ⋅ IFOV ⋅ t ∝ (3.11)
L
where NEΔR = noise equivalent differential radiance (unit: W m–2 sr –1).
The situation is different in the microwave range for two reasons. First, the need to limit the
antenna size means that diffraction law establishes a link between the IFOV and the optics
aperture L:
1.24 ⋅ H ⋅ c
IFOV = (3.12)
L ⋅ν *
where ν* = frequency = c/λ; and c = speed of light.
Thus, there is less latitude for trade-off parameters. Second, the detection mechanism is based
on comparing the scene temperature with the "system temperature", which increases with the
bandwidth. The final outcome is that the equivalent of equations 3.8, 3.10 and 3.11 in the MW
range is:
The system temperature depends on many technological factors, and increases sharply as
frequency increases. On the one hand, equation 3.13 shows that, in the case of MW, radiometric
resolution can only marginally be improved by increasing bandwidth and integration time,
since the benefit only grows with the square root. On the other hand, because of the diffraction-
limited regime, the usual way to increase SNR by increasing optics aperture is not applicable.
That is because, if the antenna diameter is increased, the IFOV is reduced commensurately (see
equation 3.12).
CHAPTER 3. REMOTE-SENSING INSTRUMENTS 57
This short review highlights the direct impact of user and mission requirements on instrument
sizing. In addition, it shows how important it is to formulate requirements in a way that leaves
room for optimization, without necessarily compromising the overall required performance. For
instance with reference to equation 3.10, it is possible to draw a number of conclusions:
(a) For a given set of instrument parameters (L, F, τ and D*), some user-driven parameters
(NEΔT, Δν, IFOV and t) can be enhanced at the expense of others. In certain cases, this can
be done at the software level during data processing on the ground. However, if all user
requirements become more demanding and no compromises are made, a larger instrument
will be necessary.
(b) The effect of NEΔT, Δν and IFOV on instrument size is linear in relation to the optics
diameter L. The effect of t (the integration time, driven by the requirement to cover a given
area in a given time) is damped by the square root. Therefore, requirements for increased
coverage and/or more frequent observation have a lesser impact than requirements for
improved spatial, spectral and radiometric resolution.
(c) Increasing the optics aperture L has a significant impact on instrument size. Since it is very
difficult to implement optical systems with F‑number = f/L < 1, an increase in L implies an
increased focal length, and therefore a volumetric growth of overall instrument optics.
For example, a reduction in the IFOV from three to two kilometres would double the
instrument mass.
In this section, Earth observation instruments are classified according to their main technical
features. The following instrument types are considered:
Most instrument types are subdivided into more detailed categories. Examples are provided to
illustrate how instrumental features can be suited to particular applications. A comprehensive
list of satellite Earth observation instruments, with their detailed descriptions, is available
through the WMO online database of space-based capabilities, available from the WMO Space
Programme website.
(a) Operates in the VIS, NIR, SWIR, MWIR and TIR bands (that is, from 0.4 to 15 µm);
(b) Discrete channels, from a few to several tens, separated by dichroics, filters or
spectrometers, with bandwidths from ~10 nm to ~1 µm;
(c) Imaging capability: continuous and contiguous sampling, with spatial resolution in the
order of one kilometre, covering a swath of several hundred to a few thousand kilometres;
(d) Scanning: generally cross-track, but sometimes multiangle, and sometimes under several
polarizations;
Depending on the spectral bands, number and bandwidth of channels, and radiometric
resolution, the application fields may include:
(a) Multi-purpose VIS/IR imagery for cloud analysis, aerosol load, sea-surface temperature, sea-
ice cover, land-surface radiative parameters, vegetation indexes, fires, and snow cover. The
extent of the spectral range is a critical instrument feature;
(b) Ocean-colour imagery, aerosol observation and vegetation classification. The number of
channels with narrow bandwidth in VIS and NIR is a critical instrument feature;
(c) Imagery with special viewing geometry, for the best observation of aerosol and cirrus,
accurate sea-surface temperature, land-surface radiative parameters including bidirectional
reflectance distribution function. The critical instrument features are the number of viewing
angles and, when available, polarizations.
Mission Multi-purpose VIS/IR imagery for cloud analysis, aerosol load, sea-surface
temperature, sea-ice cover, land-surface radiative parameters, normalized
difference vegetation index, fires, snow cover, etc.
Main features 6 channels (channel 1.6 and 3.7 are alternative), balanced VIS, NIR, SWIR, MWIR
and TIR
Scanning technique Cross-track: 2 048 pixels of 800 m sub-satellite point (SSP) with a swath of
2 900 km
Along-track: six 1.1 km lines a second
Coverage/cycle Global coverage twice a day (LW channels) or once a day (SW channels)
Resources Mass: 33 kg
Power: 27 W
Data rate: 621.3 kbps
Central wavelength Spectral interval NEΔT or SNR at specified input spectral radiance
0.630 µm 0.58–0.68 µm 9 at 0.5% albedo
Mission Multi-purpose VIS/IR imagery for cloud analysis, aerosol properties, sea- and land-
surface temperature, sea-ice cover, ocean colour, land-surface radiative parameters,
vegetation indexes, fires, snow cover, total ozone, cloud motion winds in polar
regions, etc.
Main features 36-channel VIS, NIR, SWIR, MWIR and TIR spectroradiometer
Coverage/cycle Global coverage nearly twice a day (LW channels) or once a day (SW channels)
Resolution (SSP) IFOV: 250 m (two channels), 500 m (5 channels), 1 000 m (29 channels)
Central Spectral interval NEΔT or SNR at specified input IFOV at Primary use
wavelength spectral radiance SSP
0.645 µm 0.62–0.67 µm 128 at 21.8 W m–2 sr –1 µm–1 250 m Land/Cloud/
0.858 µm 0.841–0.876 µm 201 at 24.7 W m–2 sr –1 µm–1 250 m Aerosol
boundaries
Mission Multi-purpose VIS/IR imagery for cloud analysis, aerosol load, sea-surface
temperature, land-surface radiative parameters, normalized difference vegetation
index, fires, snow cover, wind from cloud motion tracking, etc.
Main features 12 channels, balanced VIS, NIR, SWIR, MWIR and TIR
Table 3.6. Example of imager with special viewing geometry: POLDER on PARASOL
Mission Imagery with special viewing geometry, for best observation of aerosol and
cirrus, land–surface radiative parameters including bidirectional reflectance
distribution function, etc.
Resources Mass: 32 kg
Power: 50 W
Data rate: 883 kbps
Central wavelength Bandwidth No. of polarizations SNR at specified input
spectral radiance
443.5 nm 13.4 nm – 200 at 61.9 W m–2 sr –1 µm–1
(a) Spatial resolution in the range of less than one metre to several tens of metres;
(b) Wavelengths in the VIS, NIR and SWIR bands (0.4 to 3 µm) with possible extension to MWIR
and TIR;
(i) Single channel (panchromatic) with around 400 nm bandwidth (for example,
500–900 nm);
(iii) Continuous spectral range (hyperspectral); typically has 100 channels with around
10 nm bandwidth;
64 GUIDE TO INSTRUMENTS AND METHODS OF OBSERVATION - VOLUME IV
(d) Imaging capability: continuous and contiguous sampling, covering a swath ranging from a
few tens of kilometres to approximately 100 km, often addressable within a field of regard
of several hundreds of kilometres;
(e) Applicable in LEO. GEO is not excluded but not yet in use.
High-resolution optical imagers may perform a number of missions, depending on the spectral
bands, the number and bandwidth of channels, and steerable pointing capability. Those
missions include:
(a) Panchromatic imagers: surveillance, recognition, stereoscopy for digital elevation model,
etc. Critical instrument features are the resolution and the steerable pointing capability;
(b) Multispectral imagers: land observation for land use, land cover, ground water, vegetation
classification, disaster monitoring, etc. Critical instrument features are the number of
channels and the spectral coverage;
(c) Hyperspectral imagers: land observation, especially for vegetation process study, carbon
cycle, etc. Critical instrument features are the spectral resolution and the spectral coverage.
Scanning Pushbroom
technique 35 000 detector array
Swath: 17.6 km, addressable by tilting the satellite in a variety of
operating modes
Stereo capability both along-track and cross-orbits
Mission Land observation for land use, land cover, ground water, vegetation classification,
disaster monitoring, etc.
Hyperion
Satellite New Millennium Program Earth Observing – 1 (NMP EO‑1)
Mission Land observation, especially for vegetation process study, carbon cycle, etc.
Scanning Pushbroom
technique 250 pixel/line
Swath: 7.5 km
Resolution (SSP) 30 m
Resources Mass: 49 kg
Power: 51 W
Data rate: 105 Mbps
66 GUIDE TO INSTRUMENTS AND METHODS OF OBSERVATION - VOLUME IV
(a) The nadir-viewing SW spectrometer of TROPOMI operates in the ultraviolet (UV), VIS, NIR
and SWIR bands (see Table 3.10). The polarization sensitivity of TROPOMI is reduced to less
than 0.5% with a polarization scrambler;
(b) Typical spectral resolutions of SW sounders are fractions of a nanometre (see Table 3.10);
(d) Horizontal sampling is not necessarily continuous and contiguous. However, for TROPOMI
it is contiguous, thus providing imaging capability;
(e) The scanning capabilities can be from nadir-only pointing to a swath of a few thousand
kilometres. In the case of TROPOMI, the swath is 2 600 km;
(b) UV and VIS: ozone profile and total column or gross profile of a small number of other
species, such as BrO, NO2, OClO, SO2 and aerosol;
(c) UV, VIS and NIR: ozone profile and total column or gross profile of several other species,
such as BrO, ClO, H2O, HCHO, NO, NO2, NO3, O2, O4, OClO, SO2 and aerosol;
(d) UV, VIS, NIR and SWIR: ozone profile and total column or gross profile of many other
species, such as BrO, CH4, ClO, CO, CO2, H2O, HCHO, N2O, NO, NO2, NO3, O2, O4, OClO, SO2
and aerosol;
(e) NIR and SWIR, possibly complemented by MWIR and TIR: total column or gross profile of
selected species, such as CH4, CO, CO2, H2O and O2.
Tables 3.10 and 3.11 describe one example of spectral coverage from an instrument in LEO
(TROPOMI) and another from an instrument in GEO (UVN on Meteosat Third Generation).
CHAPTER 3. REMOTE-SENSING INSTRUMENTS 67
Scanning Mechanical
technique 3-axis stabilized satellite
E–W continuous
S–N stepping
Coverage/cycle European area (lat. 30°N–65°N, long. 15°W–50°E) in 60 minutes (30 minutes is also
possible)
(a) Wavelengths in the MWIR and TIR bands (3–15 µm) with a possible extension to FIR (up to
50 µm) and auxiliary channels in the VIS/NIR bands;
(b) Spectral resolution in the order of 0.1 cm–1 (very high resolution), 0.5 cm–1 (hyperspectral)
or 10 cm–1 (radiometer);
(e) Scanning capability can be from nadir-only pointing to a swath of a few thousand
kilometres;
Depending on their spectral bands and resolution, cross-nadir scanning IR sounders may be
used for atmospheric temperature and humidity profiling, and/or in atmospheric chemistry for a
number of species:
(c) Very high-resolution spectrometers that are specifically for atmospheric chemistry provide
profiles or total columns of C 2H2, C 2H6, CFC‑11, CFC‑12, CH4, ClONO2, CO, CO2, COS, H2O,
HNO3, N2O, N2O5, NO, NO2, O3, PAN, SF6, SO2 and aerosol.
Tables 3.12–3.14 set out three examples: a radiometer in GEO (Sounder on Geostationary
Operational Environmental Satellite (GOES)); a hyperspectral sounder in LEO (IASI on Metop)
and a very high-resolution spectrometer in LEO (TES‑nadir on EOS‑Aura).
GOES Sounder
Satellites GOES-8, GOES-9, GOES-10, GOES-11, GOES-12, GOES‑13, GOES‑14, GOES‑15
Scanning Mechanical
technique Biaxial
3-axis stabilized satellite
Step-and-dwell
Mission Atmospheric chemistry: profiles or total columns of C 2H2, C 2H6, CFC‑11, CFC‑12,
CH4, ClONO2, CO, CO2, COS, H2O, HNO3, N2O, N2O5, NO, NO2, O3, PAN, SF6, SO2
and aerosol
Scanning technique Cross-track mode: array of 16 detectors with a 0.53 km2 x 0.53 km2 nadir footprint
moving in 10 steps to cover a 5.3 km2 x 8.5 km2 field of view that can be pointed
everywhere within a 45°-aperture cone or an 885 km swath.
The cross-nadir mode is alternative to the limb mode.
Coverage/cycle Cross-track mode: if used full time with strategic pointing, global coverage could
be obtained for cells of ~80 km wide in 16 days (the orbital repeat cycle)
Spectral range (µm) Spectral range (cm–1) Spectral resolution NEΔT at specified scene
(unapodized) temperature
11.11–15.38 µm 650–900 cm–1 0.059 cm–1 < 1 K at 280 K
(e) Scanning: cross-track (swath in the order of 2 000 km), conical (swath in the order of
1 500 km, possibly providing single or dual polarization) or nadir-only;
Depending on their frequency, spatial resolution, and scanning mode, MW radiometers may
perform a number of missions:
(a) Multi-purpose MW imagery for precipitation, cloud liquid water and ice, precipitable
water, sea-surface temperature, sea-surface wind speed (and direction if multipolarization
is used), sea-ice cover, surface soil moisture, snow status, water equivalent, etc. Critical
instrument features are: the extension of the spectral range, from 19 GHz as a minimum
(possibly 10 GHz, or ideally 6–7 GHz) to at least 90 GHz; and conical scanning to make use
of differential polarization under conditions of a constant incidence angle;
(b) Nearly-all-weather temperature and humidity sounding, which is also relevant for
precipitation. Critical instrument features are the channels in absorption bands: O2 for
temperature (main frequency: 57 GHz) and H2O for humidity (main frequency: 183 GHz);
(c) Sea-surface salinity and volumetric soil moisture. One critical instrument feature is the
low frequency in the L band (main frequency: 1.4 GHz); this implies the use of very large
antennas (see Figure 3.7);
(d) Atmospheric correction in support of altimetry missions. Critical instrument features are
the frequency of the water-vapour 23 GHz band and its nearby windows; and the nadir
viewing, co-centred with an altimeter.
Figure 3.7. Sketch view of Soil Moisture and Ocean Salinity (SMOS) with MIRAS (left) and the
satélite de aplicaciones científicas – D (SAC‑D) with Aquarius (right). The Aquarius real
aperture antenna measures 2.5 m in diameter. The MIRAS synthetic aperture antenna is
inscribed in a circle that measures 4 m in diameter.
Mission Multi-purpose MW imagery for precipitation, cloud liquid water and ice,
precipitable water, sea-surface temperature, sea-surface wind speed, sea-ice cover,
surface soil moisture, snow status, water equivalent, etc.
f0 ± 0.217 78 QH 1.20 K
Resolution 50 km basic; may be degraded, depending on the desired accuracy for
(SSP) salinity measurements
Scanning Nadir-only viewing, associated with the Poseidon‑3 and Poseidon‑3B radar
technique altimeters
Coverage/cycle Global coverage in 1 month for 30 km average spacing, or in 10 days for
100 km average spacing
Resolution 25 km
(SSP)
Resources Mass: 27 kg
Power: 31 W
Data rate: 100 bps
(a) Scanning of the Earth’s limb: this determines vertical resolution (in the range 1–3 km), the
observed atmospheric layer (in the range 10–80 km), and spatial resolution (about 300 km
along-view);
(c) Spatial resolution, from a few tens of kilometres to a few hundreds of kilometres in the
transverse direction;
CHAPTER 3. REMOTE-SENSING INSTRUMENTS 75
Limb sounders can observe the high troposphere, stratosphere and mesosphere with high
vertical resolution, and are mainly used for atmospheric chemistry. Depending on their spectral
bands, limb sounders may track different species:
(a) SW spectrometers for a number of species, depending on the part of the spectrum covered;
for the full range (UV/VIS/NIR/SWIR), the main species are: BrO, CH4, ClO, CO, CO2, H2O,
HCHO, N2O, NO, NO2, NO3, O2, O3, O4, OClO, SO2 and aerosol;
(b) IR spectrometers for a number of species, depending on the part of the spectrum covered;
for the full range (MWIR/TIR), the main species are: C 2H2, C 2H6, CFCs (CCl4, CF4, F11, F12,
F22), CH4, ClONO2, CO, COF2, H2O, HNO3, HNO4, HOCl, N2O, N2O5, NO, NO2, O3, OCS, SF6
and aerosol;
(c) MW spectrometers for a number of species, depending on the part of the spectrum
covered; for the range 100–3 000 GHz, the main species are: BrO, ClO, CO, H2O, HCl, HCN,
HNO3, HO2, HOCl, N2O, O3, OH and SO2;
(d) Occultation sounders, tracking the Sun, the moon or the stars, for a number of species,
depending on the part of the spectrum covered; for the full range (UV/VIS/NIR/SWIR) the
main species are: H2O, NO2, NO3, O3, OClO and aerosol.
Table 3.19 shows an example of a limb sounder that uses SW occultation (Stratospheric Aerosol
and Gas Experiment (SAGE‑III) on the International Space Station (ISS) satellite).
SAGE‑III ISS Stratospheric Aerosol and Gas Experiment – III for the ISS
Satellite International Space Station
Scanning Sun and moon tracking during occultation phase, 1 km step from
technique ~10 km to ~85 km
Coverage/cycle A few tens of events per day, limited to latitudes below ~52° (orbital
inclination of the ISS)
Resources Mass: 76 kg
Power: 80 W
Data rate: 115 kbps
(a) GNSS receivers using at least two L-band frequencies around 1 180 GHz, 1 250 GHz and
1 580 GHz;
(b) Earth’s limb observation, from surface to satellite altitude during the occultation phase
of satellites from the GNSS constellations (such as GPS, GLONASS, Galileo and Compass/
Beidou);
76 GUIDE TO INSTRUMENTS AND METHODS OF OBSERVATION - VOLUME IV
(c) Directional antennas: aft-looking (for setting GNSS), forward-looking (for rising GNSS), and
toroidal (for navigation);
(d) Effective spatial resolution at around 300 km along the direction from the LEO satellite to
the occulting GNSS satellite; a few tens of kilometres in the transverse direction;
(e) Horizontal sampling limited by the daily number of occultation events: from 250 to
1 500 events by satellite, depending on the number of GNSS systems received, and on the
aft-looking/forward-looking tracking capability;
Depending on their detailed features, GNSS radio-occultation sounders can provide different
types of information:
(a) The signal-sampling time interval determines the vertical resolution of temperature,
humidity and density profiles;
(b) Measurement sensitivity to the low atmospheric layers is determined by the size of the
occultation antennas and the time-sampling technique;
(c) The number of frequencies that are used affects the accuracy of two ionospheric
measurements: total electron content and electron density profile;
(d) The number of occultation events per day depends on the number of GNSS constellations
used (GPS, GLONASS, Galileo, Beidou), the number of receiving channels for simultaneous
tracking of further GNSS satellites, and the antenna accommodation feature: only aft-
looking, only forward-looking, or both.
Table 3.20 sets out the main features of an example of a radio-occultation sounder (GNSS
Receiver for Atmospheric Sounding (GRAS)).
GNSS transmitter
(used for tracking and
clock corrections)
GNSS transmitter
(used for tracking)
Ground station
(used for tracking LEO receiver
and clock corrections)
Earth
GNSS transmitter
Atmosphere (used for tracking)
Ground station
(used for tracking)
GNSS transmitter
(used for tracking)
Main features Measuring the phase delay due to refraction during occultation
between a navigation satellite and the LEO satellite
GNSS constellation: GPS
Frequencies: L1 = 1 575.42 MHz and L2 = 1 227.6 MHz
8 receiving channels: 4 for occultation, 8 for navigation
Resources Mass: 30 kg
Power: 30 W
Data rate: 27 kbps
(a) Wavelengths in the bands of total radiation emerging from the Earth and the atmosphere
(0.2–300 µm) as well as from the fraction represented by reflected solar radiation
(0.2–4.0 µm);
(b) One broadband channel, integrating over each of the two bands, and optional narrow-
bandwidth channels in VIS and/or TIR to collect information on clouds within the IFOV;
(c) Cross-track scanning with continuous and contiguous sampling, to cover a swath of a few
thousand kilometres with a spatial resolution of the order of 10 km;
(d) Applicable in LEO and GEO; observation from the L1 Lagrange libration point also is
possible.
Broadband radiometers are designed to measure the Earth radiation budget – upward LW and
SW irradiance at the top of the atmosphere (TOA). Accuracy depends on their detailed features:
(a) The greatest possible extension of the SW end of the spectrum into the UV range, and of
the LW end of the spectrum into the FIR range, with a maximally flat response inside those
ranges;
(c) Supportive narrowband channels to collect information on clouds within the IFOV.
Tables 3.21 and 3.22 set out an example of a broadband radiometer in LEO (Clouds and the
Earth’s Radiant Energy System (CERES)) and one in GEO (Geostationary Earth Radiation Budget
(GERB)).
78 GUIDE TO INSTRUMENTS AND METHODS OF OBSERVATION - VOLUME IV
Coverage/cycle Global coverage twice a day (IR and total radiance) or once a day (SW)
Resources Mass: 25 kg
Power: 35 W
Data rate: 50.6 kbps
Channel Spectral interval Noise equivalent Absolute accuracy SNR
radiance
SW 0.32–4.0 µm 0.8 W m–2 sr –1 2.4 W m–2 sr –1 1 250
(b) Integration over the full range (total solar irradiance) and/or spectroscopy in the
0.15–3.00 µm range;
(c) Total solar irradiance is measured by absolute techniques, such as active cavity radiometers
pointing at the Sun;
Solar irradiance monitors complement broadband radiometers in order to measure the Earth
radiation budget. They also contribute to solar activity monitoring for the purpose of space
weather observation. Detailed features that affect their performance are:
(b) Their capability to provide spectral information in context in the UV/VIS/NIR/SWIR ranges.
Table 3.23 sets out the main features of an example of a solar irradiance monitor in LEO (Total
and Spectral Solar Irradiance Sensor (TSIS)).
Scanning Sun pointing during orbital movement; data sampled every two
technique minutes
Coverage/cycle 100 minutes: one measurement after integration on all data taken
during the diurnal orbit arc
(a) Detector matrix (CCD): continuous Earth observation in a very narrow O2 band at 777.4 nm;
(d) Continuous and contiguous horizontal sampling; swath of several hundred kilometres from
LEO, and full disk from GEO;
Lightning imagery is useful as a proxy for convective precipitation and turbulence, in order
to monitor the Earth's electric field, and as a proxy for NOx generation. Different sampling is
applicable from LEO and GEO:
(a) From LEO, the measurement is available for the interval during satellite motion in which one
Earth’s spot is visible within the field of view of the CCD matrix (about 90 seconds);
Tables 3.24 and 3.25 set out an example of a lightning imager in LEO (Lightning Imaging Sensor
(LIS)) and one in GEO (Geostationary Lightning Mapper (GLM)).
Main features CCD camera operating at 777.4 nm (O2) to count flashes and measure
their intensity
Resources Mass: 21 kg
Power: 33 W
Data rate: 6 kbps
Main features CCD camera operating at 777.4 nm (O2) to count flashes and measure
their intensity
Resources Mass: 35 kg
Power: 110 W
Data rate: 77 Mbps
CHAPTER 3. REMOTE-SENSING INSTRUMENTS 81
(b) Pulse repetition rate that results in a vertical resolution of a few hundred metres;
(d) Continuous and contiguous horizontal sampling; swath from nadir-only to several
hundreds of kilometres;
(a) Ku band is suitable for heavy rain (liquid, with droplets that may exceed 1 cm). Non-
precipitating clouds (droplets of less than 0.1 mm) are totally transparent, and light
precipitation can hardly be detected. At these relatively low frequencies, electronic
switching, which is necessary to avoid mechanical movements of large antennas, is
relatively easy. Relatively wide swaths (several hundreds of kilometres) can therefore be
implemented.
(b) Ka band is suitable for light rain (from stratiform clouds) and snowfall. Electronic switching
is still possible, and swaths of a few hundreds of kilometres can be implemented.
(c) W band is suitable for non-precipitating clouds (droplets of less than 0.1 mm). Several
studies have also reported that this can be applied to the observation of precipitating cloud
systems, specifically on the edges of precipitation or in cases of no precipitation, including
the eye of tropical cyclones.
Tables 3.26 and 3.27 set out an example of a dual-frequency (Ku and Ka) precipitation radar (DPR)
and an example of a W-band cloud radar (Cloud Profiling Radar (CPR) on CloudSat).
Table 3.26. Example of precipitation radar: DPR on the Global Precipitation Measurement
(GPM) Core Observatory
Mission Vertical profile of heavy rain (liquid), light rain and snowfall
Coverage/ Global coverage in 1 month for 30 km average spacing, or in 10 days for
cycle 100 km average spacing
(b) Very accurate calibration in order to measure backscatter coefficients (σ 0) from sea capillary
waves;
There are two concepts, mainly differing by the scanning principle (see Figure 3.9):
(a) Electronic scanning: side-looking, generally uses C band and provides three azimuth views
for differential σ 0. It is more accurate for low-intensity sea-surface wind and for soil moisture.
(b) Conical scanning: generally uses Ku band, with two beams and two polarizations. It
provides four azimuth views for differential σ 0.
Tables 3.28 and 3.29 describe radar scatterometers using pushbroom scanning (ASCAT) and
conical scanning (SeaWinds).
CHAPTER 3. REMOTE-SENSING INSTRUMENTS 83
Ground track
Left side Right side
SeaWinds
Outer beam
45°
Left side swath Right side swath
Mid beam Mid beam
Left side Right side
135°
550 km
500 km Instantaneous
antenna footprint Inner beam
700 km
25 km 900 km
Nadir track
Figure 3.9. Two concepts for multiviewing scatterometers. Left: six antennas for three σ = s
under azimuth angles, 45°, 90° and 135° respectively, on both the left and right side of the
sub-satellite track (Advanced Scatterometer (ASCAT) on Metop). Right: conical scanning of
an antenna with two beams and two polarizations, for σ = s under four azimuth angles for
areas in the inner circle (SeaWinds on QuikSCAT). The ASCAT concept leaves an uncovered
strip of ~700 km around the sub-satellite track. In the SeaWinds concept, there appears to be
no gap, but accuracy is poor in the inner part of the swath around the sub-satellite track.
SeaWinds
Satellite Quick Scatterometer (QuikSCAT) mission
Scanning Conical scanning: 2 beams to provide four views of each spot from
technique different angles
Swath: 1 800 km
(a) Operating frequencies in Ku band (~14 GHz), with auxiliary C band (~5 GHz), or Ka band
(~35 GHz);
(b) Very accurate ranging measurement between the satellite and the Earth’s surface;
Radar altimeters generally operate in Ku band, using C band for correction of the signal rotation
induced by the ionosphere. They are linked to a nadir-pointing MW radiometer for water vapour
correction. Their accurate ranging is used for ocean topography: the echo spread provides
information on significant wave height, while the echo intensity provides information on wind
speed.
Depending on the detailed features of the instrument and of the satellite orbit, altimeters may be
optimized for different applications:
(b) SAR-like processing of the return echoes to synthesize higher spatial resolution along the
sub-satellite track (see Figure 3.10);
(d) Dual frequency (C and Ku bands), which provides information on the total electron content
between a satellite and the Earth's surface.
CHAPTER 3. REMOTE-SENSING INSTRUMENTS 85
Table 3.30 sets out the main features of a radar altimeter with data of geodetic quality
(Poseidon‑3).
Mission Ocean topography, the geoid, significant wave height, wind speed,
total electron content
Resources Mass: 70 kg
Power: 78 W
Data rate: 22.5 kbps
(b) Several combinations of polarizations in transmission and reception: HH, VV, VV/HH,
HH/HV and VV/VH;
(c) There is a trade-off between spatial resolution and swath: a 1–30 m resolution is associated
with a swath of 30–100 km; but a 100–1 000 m resolution is associated with a swath of
300–500 km;
86 GUIDE TO INSTRUMENTS AND METHODS OF OBSERVATION - VOLUME IV
(d) Side looking, generally on one side, maintaining high resolution within a field of regard of
several hundreds of kilometres;
Figure 3.11 illustrates the operating modes of one C-band SAR (Advanced Synthetic Aperture
Radar (ASAR)).
The operating frequency is a critical feature, optimized for the purposes for which an SAR is
designed:
(b) L band is best suited to wave observation and volumetric soil moisture;
(d) C band covers a wide range of applications: sea ice, wave parameters by spectral analysis
of image segments, surface soil moisture, snow parameters, glaciers, ground water, etc.
However, each individual parameter can be optimally observed at other frequencies;
(e) X band provides the best spatial resolution, and is therefore best suited to surveillance;
(g) Interferometry of the signals from one SAR at different times or two SARs flying in tandem
enables the measurement of the digital elevation model and the detection of contour
changes (such as coastlines and lakes) and elevation (for example, volcano top surface).
irection
Flight d
485 km
100 km
Wave
VV or HH
405 km Alternating polarization < 30 m resolution
VV or VH or HH/HV 5 km x 5 km
Image vignettes
30 m resolution
Wide swath VV or HH
(PRI–GEC)
VV or HH < 30 m resolution
Global monitoring Up to 100 km swath
150 m resolution (PRI–GEC)
VV or HH
405 km swath width Up to 100 km swath
1 000 m resolution
405 km swath width
PRI = precision image GEC = geocoded image
Figure 3.11. Operating modes of ASAR on Envisat. In the global monitoring and wide swath
modes, the swath is 405 km and is linked to either a 1 000 m or 150 m resolution. In the
image and alternating polarization modes, a 100 km swath with a 30 m resolution can be
pointed to one of seven positions within a field of regard of 485 km. In the wave mode,
vignettes of 5 km x 5 km with 30 m resolution are sampled at every 100 km along track.
CHAPTER 3. REMOTE-SENSING INSTRUMENTS 87
Mission Sea ice, wave parameters by spectral analysis of image segments, surface soil
moisture, snow parameters, glaciers, ground water, etc.
Coverage/cycle Global coverage in 5 day in global monitoring mode (if used for 70% of the time)
Longer periods (up to 3 months) for other operation modes
Resolution (SSP) 30 m to 1 km, depending on operation mode (see lower part of the table)
(a) Operating wavelengths in the UV (for example, 355 nm), VIS (for example, 532 nm), NIR
(for example, 1 064 nm), or SWIR (for example, 1 600 nm) bands;
(b) Possible dual wavelength, two receivers (for Mie and Rayleigh scattering); polarimetry;
(c) Horizontal resolution within a 100 m range, often degraded by up to 50 km in order to
collect enough de-correlated samples;
(d) Vertical resolution within a 100 m range (approximately 10 cm for lidar altimeters);
A space lidar is a voluminous instrument that needs to be optimized for specific applications:
(a) Doppler lidars generally operate in UV, for both Mie and Rayleigh scattering, in order to
track aerosol and air molecules; oblique view is used to measure radial wind in clear air and
aerosol;
88 GUIDE TO INSTRUMENTS AND METHODS OF OBSERVATION - VOLUME IV
(b) Backscatter lidars operate at one (on UV) or two (VIS and NIR) wavelengths, often with
more polarizations; nadir view is used for obtaining aerosol profiles, cloud-top height and
atmospheric discontinuities, such as the height of the top of the planetary boundary layer
and of the tropopause;
(c) Lidar altimeters usually operate at two wavelengths (VIS and NIR); nadir view is used, as
is very high-vertical resolution (for sea-ice elevation) and horizontal resolution (for ice
boundaries);
(d) Differential absorption lidar operate at one wavelength centred on the absorption peak of
one trace gas, in UV, VIS, NIR or SWIR, and nearby windows; nadir view is used for high-
vertical-resolution observation of, for example, O3, H2O and CO2.
Tables 3.32–3.35 give details of a Doppler lidar (Atmospheric Laser Doppler Instrument
(ALADIN)), a backscatter lidar (Cloud-aerosol Lidar with Orthogonal Polarization (CALIOP)), a
lidar altimeter (Geoscience Laser Altimeter System (GLAS)) and a differential absorption lidar
(CO2 lidar).
Scanning No scanning
technique Pulse echoes averaged over a 50 km field of view
Field of view sampled at 200 km intervals
Coverage/ Global coverage in 1 month for 30 km average spacing, or in 10 days for
cycle 100 km average spacing
Coverage/ Global coverage in 1 month for 30 km average spacing, or in 10 days for
cycle 100 km average spacing
Mission Polar ice sheet thickness and topography, cloud-top height, aerosol
CO2 lidar
Satellite Active Sensing of CO2 Emissions over Nights, Days, and Seasons
(ASCENDS)
Coverage/cycle Global coverage in one month for 30 km average spacing, or in 10 days
for 100 km average spacing
3.2.16 Gradiometers/accelerometers
Knowledge of the gravity field is crucial for modelling the solid Earth. Several space techniques
address this subject.
(a) The long-wave components of the gravity field are measured through radar or lidar
altimetry, or through precise orbitography (for example, with laser ranging, radio
positioning, GNSS, star tracking).
(b) SW components (anomalies and perturbations of the gravity field) are observed at satellite
altitude by accelerometers or gradiometers, in association with satellite-to-satellite ranging
systems. An accelerometer measures the variation of the gravity field along the satellite
trajectory. A gradiometer comprises a network of accelerometers, which measures the
gravity-gradient tensor. Satellite-to-satellite ranging systems are transmitter-receiver
systems, usually in K band (24 GHz) and Ka band (32 GHz). They are designed to accurately
measure the distance and its variations between satellites flying in coordinated orbits. The
same measurements are possible through the simultaneous reception of signals from tens
of GNSS satellites: that determines positioning changes highly accurately.
Solar activity is monitored either by remote-sensing or in situ in the solar wind, from deep space
and Earth’s orbit. Several measurement approaches are possible:
(b) Energetic particles (electrons, protons, α‑particles, ions, cosmic rays, neutrons): the energy
range is generally broken down into high-, medium- and low-energy; the boundaries of
the ranges depend on the type of charged particle; measurements can be integrated over
the full energy range, or over partial ranges; spectroscopy within a range may also be
performed;
(c) Magnetic and electric fields, directly measured in the solar wind, and inferred in
the photosphere; those fields are inferred from measurements in the solar wind or
from spectroscopy of VIS solar images using the Zeeman effect, Doppler analysis or
multipolarization;
(d) Measurements can be performed by: integrating over the full solar disk; imaging the solar
disk; or imaging the corona only by occulting the disk (a coronagraph);
(e) A specific observation is that of solar irradiance, either total or spectrally resolved
(see 3.2.9).
An example of an instrument package for solar activity monitoring from the L1 Lagrange
libration point, the Solar and Heliospheric Observatory (SOHO), is described in Table 3.38.
Space environment monitoring at platform level provides information used for monitoring and
making predictions about overall space weather conditions, as well as for platform safety. The
instrumentation generally includes:
(a) Charged particle detectors, designed for specific ranges of energy, either integrated or
spectrally resolved;
Magnetic and electric fields in the magnetosphere can be measured in situ as the satellite moves
along its orbit. If the orbit is highly eccentric, it crosses the magnetosphere at different altitudes,
thus providing 3D profiles. Gradients of the fields are better observed when more satellites are
flown together in coordinated orbits. Usual instruments are:
SOHO instrumentation
Satellite Solar and Heliospheric Observatory (SOHO)
Resources Mass: 17 kg
Power: 13.6 W
Data rate: 16 kbps
Cluster instrumentation
Satellites Cluster A, B, C and D (four satellites flying together in coordinated
orbits)
The measurements described in the present volume, Chapter 2 are performed within satellite
programmes1 implemented by space agencies with either an operational mandate to serve
particular user communities, or a priority mandate for research and development. In addition to
the core meteorological constellations in geostationary and near-polar Sun-synchronous orbits,
these programmes include environmental missions focusing on specific atmospheric parameters,
ocean and ice, land observation, solid Earth or space weather. Many of these environmental
missions are designed and operated in a research or demonstration context, but some of them
have reached operational maturity, and contribute to the sustained observation of environmental
components, especially when they have been extended over time and/or they give way to an
operational follow-on. Further information on measurement principles and uncertainties for
geophysical variables can be found in the present volume, Chapter 5.
For each type of application, satellite missions may be seen as constituent parts of constellations
of spacecraft that, in many cases, will deliver their full benefit only when implemented in a
coordinated fashion, ensuring synergy among the different sensors. International coordination
among satellite operators is achieved within the Coordination Group for Meteorological
Satellites, the primary goal of which is to maintain the operational meteorological and
climate monitoring constellations, and the Committee on Earth Observation Satellites, which
has initiated “virtual constellations” with thematic objectives (ocean surface topography,
precipitation, atmospheric composition, land surface imaging, ocean surface vector wind, ocean
colour radiometry and sea-surface temperature).
The system of operational meteorological satellites constitutes the backbone of the space-based
GOS. It is split into two components, according to the orbital characteristics:
The geostationary orbit is particularly suited for operational meteorology because it enables very
frequent sampling (at sub-hourly or minute rates) as necessary for rapidly evolving phenomena
(daily weather) or detecting events such as lightning, as long as no very high spatial resolution is
required (order of 1 km). The primary observations from the geostationary orbit are:
1
Details on these programmes are available in the WMO online database on space-based capabilities, which is
updated on a regular basis.
CHAPTER 4. SATELLITE PROGRAMMES 95
(a) Cloud evolution (detection, cover, top height and temperature, type, water phase at cloud
top, particle size);
(c) Winds by tracking clouds and water-vapour patterns (including wind profile from water-
vapour profile tracking);
(d) Convective precipitation (in combination with MW data from LEO satellites and lightning
detection);
(e) Rapidly changing surface variables (sea-surface temperature in coastal zones, fires);
(f) Ozone and other trace gases affected by diurnal variation or arising from changing sources.
One drawback of the geostationary orbit is the poor visibility at high latitudes, beyond about
60° for quantitative measurements and 70° for qualitative. This limitation can be overcome
by using high-eccentricity inclined orbits (Molniya, Tundra or three-apogee orbits) instead of
the geostationary orbit (see the present volume, Chapter 2, 2.1.4). Additionally, the diffraction
limit due to the small angles subtended by the large distance poses challenges for very high-
resolution optical imagery and MW radiometry. MW observation for all-weather temperature
and humidity sounding and quantitative precipitation measurement from GEO should be
feasible by using high frequencies, as the technology becomes available.
The requirement for global non-polar frequent observations from the geostationary satellites
calls for six regularly-spaced spacecraft (Figure 4.1). In an operational constellation, backup
satellites are required to provide redundancy above this minimum.
Table 4.1 lists the operational programmes that agreed to contribute to the constellation of
meteorological geostationary satellites in 2012, and their nominal positions. Other positions may
be used on a temporary basis, for instance in contingency situations.
60°N
30°N
30°S
60°S
150°W 120°W 90°W 60°W 30°W 0 30°E 60°E 90°E 120°E 150°E
Figure 4.1. Coverage from six regularly-spaced geostationary satellites. The circles
subtend a geocentric angle of 60°, considered the practical limit for quantitative observations
(for qualitative use, images actually extend beyond). All latitudes between
55°S and 55°N are covered.
96 GUIDE TO INSTRUMENTS AND METHODS OF OBSERVATION - VOLUME IV
INSAT and Indian National Satellite and Kalpana ISRO 74°E and 93.5°E
Kalpana
FY‑2 and FY‑4 Feng‑Yun‑2 and follow-on Feng‑Yun‑4 CMA 86.5°E and 105°E
The Sun-synchronous orbit provides global coverage necessary for applications such as
global numerical weather prediction (NWP), polar meteorology, and climatology. For these
applications, very frequent sampling is less critical than global coverage and high accuracy. The
primary contributions from Sun-synchronous orbits are:
(d) Surface variables (sea and land surface temperatures, vegetation and soil-moisture
indexes);
(g) Ozone and other trace gases for environment and climate monitoring.
Additional advantages of Sun-synchronous and other low Earth orbits are the capability to
perform active sensing in the MW (radar) and optical (lidar) ranges and limb measurements of
the higher atmosphere. Global coverage at roughly four-hour intervals can be achieved by three
Sun-synchronous satellites in coordinated orbital planes crossing the equator at, for instance,
05:30, 09:30 and 13:30 local solar time, provided that the instrument swath is sufficiently wide
and the measurement can be performed by both day and night (see Figure 4.2). Table 4.2
lists the operational programmes contributing now or in the future to the constellation of
meteorological Sun-synchronous satellites as of 2012.
CHAPTER 4. SATELLITE PROGRAMMES 97
Figure 4.2. Coverage from three Sun-synchronous satellites of height 833 km and equatorial
crossing time (ECT) regularly spaced at 05:30 d, 09:30 d and 13:30 a. For the purpose of this
schematic diagram, all satellites are assumed to cross the equator at 1200 universal
coordinated time (UTC). The figure refers to a time window of 3 h and 23 min (to capture two
full orbits of each satellite) centred on 1200 UTC. Three typical swaths are considered. Nearly
three-hour global coverage is provided for the VIS/IR imagery mission, whereas for the IR/
MW sounding mission coverage is nearly complete at latitudes above 30 degrees. For
microwave conical scanners, global coverage in three hours would require eight satellites.
4.2.1 Precipitation
Precipitation is a basic meteorological variable, but its measurement requires the exploitation of
the microwave spectral range at a resolution consistent with the scale of the phenomenon and at
relatively low frequencies; this implies large instruments. Moreover, the relation between passive
MW sensing and precipitation is not explicit. Only total-column precipitation is measured, and
only in a few channels. The retrieval problem is strongly ill-conditioned and requires modelling
of the vertical cloud structure, which can only be observed by radar. TRMM (launched in 1997),
that carries associated passive and active MW sensors, has enabled algorithms to be developed
that allow much better use of passive measurements.
TRMM has enabled the concept of the GPM mission to be developed that is being implemented
in an international context. Its objective is to provide global coverage of precipitation
measurements at three-hour intervals. Since the baseline instrument is an MW conical scanning
radiometer with limited swath, the three-hour frequency requires eight satellites in regularly
distributed near-polar orbits (Figure 4.3). In addition to those “constellation satellites”, a “Core
Observatory” in inclined orbit equipped with precipitation radar enables all other measurements
from passive MW radiometers to be “calibrated” when constellation and core-satellite orbits
cross one another. Beyond the missions specifically tailored for precipitation observation, any
operational mission equipped with MW radiometers can contribute to the composite system.
Radio occultation of GNSS satellites is a powerful technique for providing temperature and
humidity profiles with a vertical resolution that is unachievable by nadir-viewing instruments.
However, the implementation of operational systems is proceeding slowly. One difficulty is
that the payload, although of low mass, power and data rate (see, for instance, the description
of GRAS in the present volume, Chapter 3, 3.2.7, Table 3.23), places volumetric constraints
on the platform (two large antennas, say 0.5 m2 each, requiring unobstructed view fore
and aft). Another difficulty is that the technique requires a significant number of satellites in
different orbits.
The radio occultation concept was demonstrated in space in 1995 by the Global Positioning
System/Meteorology (GPS/MET) on MicroLab‑1. Since then, establishing a constellation of
radio occultation receivers has been advocated, initially for climatological purposes to provide
“absolute” measurements that can be compared at any time intervals to detect climate
trends, then also for NWP high vertical resolution soundings and for the absolute reference
measurements correcting the biases of other sounding systems.
Radio occultation is an infrequent event. By exploiting one GNSS constellation and tracking
both rising and setting occultations, about 500 occultation events per day may be captured. In
addition to the long-standing GPS and GLONASS, a third constellation, Compass (named Beidou
in Chinese), is now operated by China, and a fourth constellation, Galileo, is being implemented
by the European Commission (EC) and the European Space Agency (ESA). The number of
occultations per day per satellite rises to 1 000 by exploiting two constellations and 1 500 with
three constellations if received in both fore and aft views. It has been estimated that, in order to
provide global coverage with an average sampling of 300 km every 12 h, it is necessary to deploy
at least 12 satellites on properly distributed orbital planes. One very effective approach is to
use clusters of small dedicated satellites placed in orbit by a single launch. COSMIC includes six
micro-satellites launched at once and thereafter separated into regularly-spaced orbits. Several
meteorological satellites are also carrying individual GNSS radio-occultation receivers.
A limitation of NWP and general circulation models is the representation of the radiative
processes in the atmosphere. Aerosols, cloud interior (particularly ice), radiation fluxes within
the 3D atmosphere in addition to TOA and the Earth surface, are their main factors. Some of
these variables require large observing instruments (lidar, cloud radar, etc.) that are not feasible
for multi-purpose operational meteorological satellites, thus the observation of atmospheric
radiation relies on a suite of instruments flown either in operational programmes or on dedicated
missions.
Atmospheric radiation was the first observation performed from space in October 1959 on
Explorer VII. At the time of the first TIROS flights the Earth’s planetary albedo was poorly known.
Instruments exploiting multi-viewing, multi-polarization and multi-spectral sensing have been
developed, the first one being the Polarization and Directionality of the Earth’s Reflectances
(POLDER) on the Advanced Earth Observing Satellite – 1 (ADEOS‑1; 1996–1997).
Atmospheric chemistry has grown in importance over time. Attention was focused initially on
ozone monitoring, especially after the discovery of the ozone hole; then on the greenhouse effect
as a driver of global warming; and finally on air quality, for its impact on living conditions in the
biosphere. Depending on the objective, the instrumentation may be quite simple (such as for
total column of one or few species) or very complicated (such as for vertical profiles of families of
species).
100 GUIDE TO INSTRUMENTS AND METHODS OF OBSERVATION - VOLUME IV
The A-Train
Aqua
CloudSat
CALIPSO
Glory PARASOL
Aura OCO
Figure 4.4. The spread of equatorial crossing times across the satellites addressing
atmospheric radiation (Glory, PARASOL, CALIPSO, CloudSat and EOS‑Aqua) is about two
minutes. Note that there may be some changes in the satellites participating in the A‑Train;
for instance PARASOL was removed after five years, EOS‑Aura has been added, Glory failed at
launch, the Orbiting Carbon Observatory (OCO), lost at launch, will be replaced by OCO‑2,
and the Global Change Observation Mission for Water - 1 (GCOM‑W1) has been added.
It is noted that:
The first comprehensive mission for atmospheric chemistry, the UARS, was exploiting limb
sounding. When launched in 1991 it was by far the largest Earth observation satellite ever in orbit
(mass at launch: 6 540 kg). Table 4.3 provides a list of satellites either substantially addressing, or
fully dedicated to, atmospheric chemistry.
The lack of limb sounding on future satellites jeopardizes the observation of the higher
atmosphere.
The study of atmospheric dynamics involves missions measuring the 3D wind field, a difficult
issue since the wind per se does not have a signature in the electromagnetic spectrum.
Nevertheless, considerable effort has been devoted and continues to be devoted to the subject,
since wind is a primary observation for NWP and general circulation models.
Deriving wind characteristics from the motion of clouds or other atmospheric patterns has been
an early application of geostationary satellites. Still today, it is the operational practice providing
thousands of wind vectors every day. Tracking clouds or water vapour patterns, however,
determines the wind at one level only. The level depends on the tracer and is measured with
limited accuracy. For large areas, all tracers tend to be in the same altitude range, thus limiting
the vertical resolution in practice to one or two levels. With the future advent of hyperspectral
sounders in GEO, frequent water vapour profiles, with their patterns, will be available at
CHAPTER 4. SATELLITE PROGRAMMES 101
Table 4.3. Satellite programmes with strong or exclusive focus on atmospheric chemistry
EOS‑Aura Earth Observation System – Aura NASA IR limb, IR nadir and limb,
UV/VIS nadir, MW limb
additional heights, and a broad vertical resolution will be achieved in clear air. Atmospheric
motion winds are also derived over polar areas from polar-orbiting satellites, taking advantage of
the frequent overpasses.
Experiments have been conducted to demonstrate the tracking of atmospheric eddies, aerosol
and molecules by Doppler lidar, capable of very high vertical resolution in clear air. This is the
scope of the ADM – ALADIN.
Winds are also of interest in the stratosphere and mesosphere, where clouds and water vapour
have no characteristic pattern, and there are no turbulence eddies or dense aerosols either. The
technique applicable here is measurement of the Doppler shift of narrow lines in the oxygen
band around 760 nm. Demonstrated by UARS, the technique is exploited by the Thermosphere,
Ionosphere, Mesosphere Energetics and Dynamics mission (TIMED) with TIDI (TIMED Doppler
Interferometer).
Certain observations of ocean and sea ice have been provided by meteorological satellites since
the very beginning of the space era. VIS imagery, the very first application of meteorological
satellites, is capable of sea-ice mapping. IR imagery added the capability to measure sea-
surface temperature. MW imagery extended the observing capability to measure sea-surface
temperature and ice cover to all-weather conditions, and added the capability to sea-surface
wind speed. Radar scatterometry started in 1978. These observations of sea-surface temperature,
102 GUIDE TO INSTRUMENTS AND METHODS OF OBSERVATION - VOLUME IV
sea-surface wind, and ice cover are still provided by operational meteorological satellites. Further
measurements, including altimetry, ocean colour, salinity, and waves, are performed in the
framework of non-meteorological programmes, sometimes dedicated to ocean and sea ice.
Table 4.4. Satellite programmes addressing missions on ocean and sea ice
ICESat Ice, Cloud and Land Elevation NASA Lidar altimetry for ice
Satellite
SARAL Satellite with Argos and Altika CNES, ISRO Radar altimetry
CHAPTER 4. SATELLITE PROGRAMMES 103
SMOS Soil Moisture and Ocean Salinity ESA Ocean salinity (synthetic
aperture antenna)
Ocean surface topography is a primary mission for oceanography, since it provides insight into
the large-scale ocean circulation that is a major oceanographic feature and a basic component of
the climate system. Reconstruction of ocean topography also implies accurate determination of
the geoid, an insight into the solid Earth domain.
Radar altimetry is the only observing technique for ocean topography. Altimeters are carried by
different satellites, some multi-purpose platforms to provide global coverage, and some (since
the Topography Experiment – Poseidon in 1992–2006) dedicated satellites in specially stable
orbits to provide accurate reference measurements. Some altimeters are specifically designed
for polar-ice topography, in one case exploiting lidar for more accurate boundary detection and
vertical resolution (see Table 4.4).
Ocean colour is a very informative observation to infer the state of health of the ocean, its
productivity and its capacity to interact with the atmosphere, as a CO2 sink, for example.
Therefore, it is a basic observation for operations in open ocean and in coastal zones, and for
climatology.
Several satellites address ocean colour observation. Some of them belong to the operational
meteorology framework; one is in GEO (see Table 4.4).
Wind over the sea surface is a basic measurement for oceanography since it drives atmospheric
forcing, hence surface currents, and the intensity of air–sea interaction. Of course, it is also an
important geophysical variable for weather prediction, enabling derivation of surface pressure
that cannot be directly measured from space. Therefore, several wind-observing instruments are
part of the operational meteorological mission. Table 4.4 records the missions addressing sea-
surface winds that are able to provide both speed and direction (radar scatterometer either in the
C band operated in the pushbroom manner, which involves scanning a swath straight down, or
in the Ku band with conical scanning; and polarimetric MW passive radiometers). Other passive
MW radiometers, that can contribute to wind-speed observation, are mentioned only for multi-
purpose MW imagers with large antennas.
104 GUIDE TO INSTRUMENTS AND METHODS OF OBSERVATION - VOLUME IV
Salinity is a basic measurement for oceanography since, with temperature, it controls water
density, hence the vertical motion in the thermo-haline layer. Furthermore, it controls the ocean’s
ability to remove trace gases from the atmosphere. Measuring salinity from space is possible
only by low-frequency MW radiometry (L band around 1.4 GHz), which requires large antennas.
Missions addressing sea-surface salinity are based either on a synthetic-aperture antenna, or on a
real-aperture antenna (see Table 4.4).
4.3.5 Waves
The observation of waves is important for ocean operations, in the open ocean and even more
so in coastal zones. It is also important for coastal-zone climatology. Unfortunately, waves are
difficult to observe from space, since the direct measurement from the radar altimeter provides
only the significant wave height, and only along the satellite track. The altimeter missions are
listed in Table 4.4.
The 2D wave field can be observed by spectral analysis of SAR imagery. In principle, any vignette
from a SAR image could be processed to provide the dominant wave direction and period, as
well as the directional energy frequency spectrum. In practice, the vignettes are sampled at
intervals during the whole orbit, and stored on board since the associated data rate is rather low.
The Envisat ASAR performs this function in wave mode.
All imagery missions of the operational meteorological satellites provide information on several
geophysical variables characterizing land surface. Specifically:
(a) VIS/IR imaging instruments: land surface temperature, soil moisture indexes, several
vegetation indexes, several fire parameters, radiative parameters, ice and snow cover;
(c) Radar scatterometers: surface-soil moisture, total biomass, snow water equivalent.
However, the design of the instruments flown on operational meteorological satellites is driven
by the main objective of describing the surface–atmosphere interface processes, as necessary
(and sufficient) for weather analysis and prediction and by the need for the observed spatial-
temporal scales to be consistent with climate monitoring requirements. This section focuses
on satellite programmes addressing land applications as primary objectives, for geophysical
variables such as land cover and use, fraction of vegetated land, vegetation type, lake and glacier
cover, topography, small-scale soil moisture and snow parameters for hydrology.
These applications require spatial resolutions at the scale of metres or a few tens of metres,
which imply using optical bands (especially VIS) or imaging radar (SAR). Another application of
very high-resolution optical imagery or SAR is security, including disaster monitoring, control of
compliance with internationally agreed protocols for environmental protection, etc.
Land observation has been the second space application, after meteorology, to give rise to
operational programmes. The first land-observation satellite, initially named the Earth Resources
Technology Satellite, thereafter re-named Landsat‑1, was launched by NASA in July 1972. Since
CHAPTER 4. SATELLITE PROGRAMMES 105
CBERS China-Brazil Earth Resources Satellite CAST, INPE Earth resources search and
management
Landsat and Landsat and Landsat Data Continuity USGS, NASA Earth resources search and
LDCM Mission management
ResourceSat Satellite for Earth Resources ISRO Earth resources search and
management
SPOT 4, 5 Satellite Pour l’Observation de la Terre CNES Earth resources search and
management
then, other space agencies have undertaken land-observation programmes, often in a fairly
operational way. Table 4.5 lists the programmes that have a demonstrated heritage of continuity
or are designed for long-term continuity.
The Disaster Monitoring Constellation (DMC) initiative was originally promoted by the British
National Space Centre (BNSC). The principle of DMC (see Figure 4.5) was to have about five
satellites in the same orbit, separated by about 20 min, in such a way that the (narrow) swath
of the instruments of one satellite (~600 km) was contiguous with the next, thus ensuring
daily global coverage. The first DMC satellite was AlSat‑1, launched in November 2002. A
cluster of three satellites, UK‑DMC‑1, NigeriaSat‑1 and Bilten Satellite, was placed in orbit by a
single launch in September 2003. Over time, more partners joined the constellation, and the
instrumentation became more elaborate. Table 4.6 depicts the current situation, which includes
satellites not strictly part of the DMC, but close in objective, structure and instrumentation.
106 GUIDE TO INSTRUMENTS AND METHODS OF OBSERVATION - VOLUME IV
Figure 4.5. Concept of the DMC, based on five satellites dephased by about 20 min
All missions listed for land observation have a common limitation: observations are not available
in the presence of clouds. In most cases, night-time is also a limitation since most instruments
use only the VIS spectral range. In an emergency, when high resolution is needed, it is very
important to have an all-weather sensing capability, which can be provided only by SAR. Several
SAR missions are available, many of them being managed with a perspective for long-term
operational continuity.
The number of SARs in orbit is important, since SAR instruments have a narrow swath, whereas
their application to disaster monitoring requires frequent revisit times. In addition, the SAR
principle is applicable with a single frequency, whereas various features to be observed
have “signatures” at different frequencies. The SIR‑C/X‑SAR mission (Shuttle Imaging Radar
with Payload C/X‑SAR), flown twice on the US Space Shuttle in April and September 1994,
demonstrated the benefit of having simultaneous SAR imagery in the L, C and X bands (L and C
provided by NASA, X by DLR and ASI).
Table 4.7 lists all current and planned missions equipped with an SAR, grouped by frequency
band and responsible agency.
Since the early days of space missions, satellites have been used to reconstruct the shape of the
geoid by means of various orbits of different heights, inclinations and eccentricity. The main
purpose was internal: to support mission analysis for orbiting satellites. As time passed and
technology has improved, the purpose has evolved towards the study of the Earth itself.
The satellite objectives for solid Earth are (see Figure 4.6 for definitions):
(a) To provide a very accurate determination of the geoid, which is the basis for several
associated applications, particularly conversion of altimeter measurements into sea level
and ocean topography. The most common technique uses radar altimetry from orbits of
relatively high altitude and high stability;
(b) To infer crustal dynamics by monitoring local site positions from satellites in well-known
and stable orbits; common techniques are laser ranging and ground-based GPS receivers;
108 GUIDE TO INSTRUMENTS AND METHODS OF OBSERVATION - VOLUME IV
Paz (SEOSAR) Paz (Spanish Earth Observation SAR) CDTI, Spain X band
(c) To infer the dynamics of the outermost layers of the Earth (lithosphere, mantle, upper part
of the mesosphere); common techniques use measurements of the gravity field and its
anomalies by very low-orbiting satellites, and satellite-to-satellite tracking;
(d) To collect information on the inner parts of the globe (lower mesosphere, liquid core, solid
core) inferred through observation of the magnetosphere in satellite measurements of the
magnetic and electric fields.
(a) Those referring to the geoid and crustal positioning and movements (space geodesy);
(b) Those referring to the lithosphere and inner layers (Earth interior).
Table 4.8 lists missions specific to solid Earth, for either geodesy or the interior.
The primary technique for reconstructing the geoid (the equipotential surface which would
coincide exactly with the mean ocean equilibrium surface, if the oceans were at rest and
extended through the continents) is radar altimetry. Other information on the geoid stems from
the precise positioning systems on any satellite. Several of them are worthy of mention:
CHAPTER 4. SATELLITE PROGRAMMES 109
Lithosphere
Mantle
Mesosphere
Liquid core
Solid core
GRACE Gravity Recovery and Climate Experiment NASA, DLR, Accelerometer, laser
(2 sats) CNES ranging, satellite-to-
satellite ranging
SWARM The Earth’s Magnetic Field and Environment ESA, CNES, CSA Accelerometer,
(3 sats) Explorers electric field sensor,
magnetometers
Note:
DNSC = Danish National Space Center
(a) Laser retro-reflectors, to accurately measure the distance and rate of change for the satellite
from the laser source on the ground;
(c) Two-way and dual-frequency microwave tracking system for ground receiving stations;
(d) GNSS;
110 GUIDE TO INSTRUMENTS AND METHODS OF OBSERVATION - VOLUME IV
(e) Star trackers, utilized for satellite attitude control but also contributing to precise
orbitography.
The primary purpose of these systems is to support precise orbit determination as necessary for
certain instruments performing the satellite mission, the most sensitive of which are altimeters
and limb sounders. The benefit for the geodetic mission stems from the statistical analysis of the
data. In this section the focus is on the application with the opposite objective: to establish the
position of a ground station assuming that the orbit is well known. To this effect, an International
Terrestrial Reference System for space geodesy has been established to collect and analyse data
in a number of coordinated centres. The system includes a few satellites having space geodesy as
a unique objective. These are listed in Table 4.8.
The representation of the geoid is now fairly precise, despite its complexity. With the help of
mathematical models using spherical harmonics, the achieved accuracy is now in the range
of 1 cm or less. Figure 4.7 shows a current view of the geoid. It may be observed that, in this
representation, the Earth’s surface is not at all a regular ellipsoid, although the vertical range of
the geoid height is contained within 200 m. The regularity of the geoid is affected by undulations
of different wavelengths ranging from many thousands of kilometres to a few hundred. One
objective is to associate these anomalies with the Earth’s interior, first of all to the lithosphere
because of its relevance to volcanism and earthquakes.
Specific missions for studying the Earth’s interior exploit gravity and gravity-gradient
observations, representative of the external layers (lithosphere, mantle and upper mesosphere);
and magnetic and electric fields, significant for the internal layers (lower mesosphere, liquid core
and solid core) (see Figure 4.6). Table 4.8 lists missions on the Earth’s interior.
Although the term “space weather” is relatively recent, the relevant activities started with the
advent of the space era, if not before, because space weather has a strong impact on the safety
of satellites in orbit and of man in space. Awareness and prediction of the space environment has
now become a prerequisite for the long-term sustainability of space activities. In addition, there
is increasing awareness of the impact of space weather on facilities on the Earth.
Space weather is characterized by electromagnetic bursts from the X to the radio bands, solar
energetic particle events, and perturbations in the solar wind density and speed, such as
interplanetary coronal mass ejections propagating plasma blobs. In particular, the solar wind
modulations compress and shape the magnetosphere, and this effect propagates lower down
to the thermosphere and ionosphere. Telecommunications and even power grids, pipelines and
other conducting networks on the Earth’s surface are affected (by geomagnetically induced
currents, for example). Rapid magnetic changes on the ground, that occur during geomagnetic
storms and are associated with space weather, can also be important for activities such as
geophysical mapping and hydrocarbon production. Correlations have been discovered between
travelling ionospheric disturbances and atmospheric gravity waves in the thermosphere.
Monitoring space weather implies two main aspects: monitoring the electromagnetic and
particle solar emission as well as the solar wind to characterize the modulation source (solar
activity), and monitoring the effects of this activity within the magnetosphere and down to the
Earth’s surface.
There have been space missions to understand solar physics since the early days of the space era,
either from deep space orbits or from Earth orbits.
Two “sentinels” of solar wind, the joint NASA/ESA SOHO mission and the NASA Advanced
Composition Explorer (ACE) mission, were launched in 1995 and 1998, respectively. SOHO and
ACE have been placed at the L1 Lagrangian point (at 1% of the Earth–Sun distance upstream
of the Earth). From that vantage point the two satellites measure solar wind and the associated
magnetic field approximately one hour before they reach the Earth. In 2006, in collaboration
with several European scientific institutes, NASA launched the Solar–Terrestrial Relations
Observatory, two satellites moving in the Earth’s orbit around the Sun, viewing the Sun from
changing positions to obtain a stereoscopic view of the dynamics of coronal mass ejection and, at
the same time, to measure the local features, at the satellite position, of the solar wind.
Several missions in Earth orbit are also carrying instruments dedicated to continuous monitoring
of solar activity. Table 4.9 lists the satellites monitoring solar activity from positions in deep space
or in Earth orbit. In addition, some geostationary meteorological satellites (GOES or FY‑4 series)
contribute or will contribute to solar monitoring.
Closer to the Earth (see Figure 4.8(a) and (b)), the thermosphere and ionosphere are the
layers where space weather is more turbulent. The main driver of the ionization state of the
ionosphere is solar electromagnetic radiation (extreme ultraviolet and ultraviolet) which, in
turn, is modulated by solar activity. The ionosphere is affected by waves, storms and travelling
disturbances. Through interaction with magnetic storms, energetic particles and electrical
currents can occur, which affect radio propagation. Mapping electron density in the ionospheric
"E-region" enables ionospheric conductivity and currents to be inferred. When associated with
magnetic field data, this information enables the internal component of the magnetic field (due
to the solid Earth) to be discriminated from external components. Small-scale irregularities and
eddies of the ionosphere can cause scattering of radio waves (scintillation), which affects the
reliability of radio links crossing the ionosphere.
112 GUIDE TO INSTRUMENTS AND METHODS OF OBSERVATION - VOLUME IV
THEMIS is a NASA mission launched in 2007. It consists of a constellation of five small satellites
in highly eccentric orbits, crossing the magnetosphere at several altitudes (see Figure 4.9), with
periods ranging from 0.8 to 4 days.
(a)
Exosphere (b)
600 km
Ionosphere 300 F2
Thermosphere F
300 km F F1
200
Height (km) E E
E
85 km 100 D
Mesosphere
45 km
Stratosphere 0
12 km
Troposphere Night Day
300 600 900 1 200 1 500 10 4
10 5
10
6
Figure 4.8. (a) Atmospheric stratification below and above the mesopause; (b) Layers of
denser electronic content. The densest layer is F2, present day and night.
PROBA 1 and 2 Project for On-board Autonomy 1 and 2 ESA LEO, Sun-synchronous
THEMIS measures the magnetic field, electric fields and charged particles in order to address the
physical processes in near-Earth space that initiate the violent eruptions of the aurora occurring
during substorms in the Earth's magnetosphere. The system also includes a number of ground
stations, to detect auroras and measure the surface magnetic field.
The MMS mission developed by NASA is based on a constellation of four satellites with highly
eccentric orbits spread across the magnetosphere, similarly to THEMIS (see Figure 4.10). Plasma
analysers, energetic particle detectors, magnetometers, and electric field instruments are used
to study the microphysics of magnetic reconnection, the ultimate driver of space weather.
Table 4.10 lists a number of missions specifically addressing the magnetosphere.
Section 4.5 describes a number of missions in lower orbit that also carry instruments relevant to
the magnetosphere:
(d) SWARM: Absolute Scalar Magnetometer, Vector Field Magnetometer and Electric Field
Instrument.
Figure 4.9. The orbits of the five THEMIS satellites in the magnetosphere. The white flash
represents the energy released by a magnetospheric substorm.
114 GUIDE TO INSTRUMENTS AND METHODS OF OBSERVATION - VOLUME IV
With the advent of radio occultation sounding, the profile of the electron density across the
ionosphere has become the best measurable geophysical variable associated with space weather.
The signal from navigation satellites (GPS, GLONASS, Compass, Galileo) is affected by the
rotation of the electric field and the delay induced by the ionosphere. In order to correct for this
effect, at least two frequencies are used (now shifting to three): ~1 180 GHz, ~1 580 GHz and,
possibly, ~1 250 GHz. Differentiating the two (or three) signals yields information on:
The total electron content, although integrated along-view, is measured for changing tangent
heights; it is therefore possible to reconstruct the vertical profile by tomography. Several radio
occultation payloads are being flown, on both multi-purpose satellites and dedicated facilities
(such as COSMIC).
Radar altimeters also provide total electron content observations by exploiting two frequencies,
generally ~13.5 GHz and ~5.3 GHz. The coverage is only at nadir and tomography is not possible;
however, since altimetry missions are often orbited at high altitude (1 336 km for JASON, for
example), the measurement includes the lower part of the plasmasphere (the layer above the
thermosphere, from ~1 000 to ~40 000 km altitude).
Total electron content can also be measured directly by phase-delay analysis of the two or three
frequencies transmitted by a GNSS satellite and received by a LEO satellite. In this case, total
electron content is observed along the path from the GNSS satellite (orbit altitude ~20 000 km)
to the LEO satellite (orbit altitude ~800 km), thus in the medium plasmasphere. The number of
available GNSS satellites is rather large: ~24 each for GPS and GLONASS systems, ~30 for Galileo,
~35 for Compass, with a total close to 110 and a fair global distribution.
Figure 4.10. The four MMS satellites flying in formation. Tetrahedral pattern is used to
capture the 3D structure of the encountered reconnection sites.
CHAPTER 4. SATELLITE PROGRAMMES 115
THEMIS Time History of Events and NASA Highly elliptic, low inclination, apogees
(5 sats) Macroscale Interactions at five different altitudes
during Substorms
VAP (2 sats) Van Allen Probe (formerly NASA Highly elliptic, low inclination, crossing
RBSP, Radiation Belt Storm the radiation belts
Probes mission)
Table 4.11 presents the information available from operational meteorological satellite series
related to space weather. Radio occultation payloads (see section 4.2.2) are omitted.
116 GUIDE TO INSTRUMENTS AND METHODS OF OBSERVATION - VOLUME IV
GOES R, S, T, U Space Environment In Situ Suite (SEISS) for charged particles in solar wind and cosmic rays
Extreme Ultraviolet Sensor/X-Ray Sensor Irradiance Sensors (EXIS)
Solar Ultraviolet Imager (SUVI)
Magnetometer (MAG)
Electro‑L GGAK‑E: Heliogeophysical Instrument Complex for charged particles of solar wind and
Electro‑M cosmic rays
JPSS SEM for National Polar-orbiting Operational Environmental Satellite System (SEM‑N):
including a spectrometer for precipitating electrons and ions, a spectrometer for
medium-energy particles, and omni-directional detectors for high-energy particles
DMSP F16 to Special Sensor Ion and Electron Scintillation Monitor (SSIES)
S20 Special Sensor Precipitating Electron and Ion Spectrometer (SSJ5)
Special Sensor Magnetometer (SSM)
Special Sensor Ultraviolet Limb Imager (SSULI)
Special Sensor Ultraviolet Spectrographic Imager (SSUSI)
5.1 INTRODUCTION
This chapter offers an overview of the geophysical variables that can be observed from space and
of the performance that can be expected for their derivation. The performance is estimated by
taking into account the physical principle involved in each measurement technique and the state-
of-the-art instrument technology at the time of writing this document and in the foreseeable
future.
Assumptions are made to provide the most representative estimation in each case. The figures do
not necessarily represent the actual performance of a particular instrument, but are intended to
illustrate the relative performances of the various remote-sensing techniques.
For the purpose of the present Guide, the discussion is limited to geophysical variables that can
be retrieved by processing the output from a single instrument or a set of closely associated
instruments. Product derivation may involve complex algorithms, physical or statistical models,
and supporting information from external sources, either ancillary (necessary for processing)
or auxiliary (to help processing). The present chapter focuses on products that can be derived
with a limited amount of external information, where this external information only plays a
minor role compared with that of the satellite instrument output, and no significant bias can
be introduced by a model. For instance, modelling of the physical phenomenon controlling
the variable, radiative transfer models, and inversion retrieval models, are within the scope of
this chapter. Beyond the scope of this chapter, for example, are assimilations that merge several
measurements and background fields, that combine the physics of the phenomenon and the
dynamics of the model to the point where the satellite contribution to the output product is
barely recognizable, and that can be biased to the model being used.
This chapter will focus on Level 2 products, and some Level 3 and Level 4 products for which
there is a well-established and recognized methodology (see the processing levels defined in the
present volume, Chapter 2, 2.3.2.6, Table 2.11).
For satellite imagery used directly for human interpretation, several quality criteria can be
considered; these include spatial resolution, geo-location accuracy, calibration stability across
consecutive images, and colour constancy in representing a given property within the observed
scene in the case of RGB composite imagery. These components of the image product quality are
not discussed further here.
In this chapter, the aim is to address the quality of quantitative products with numbers that can
be used in automatic procedures and numerical models. This evaluation can then be compared
with the requirements for the same products.
User requirements may differ according to the layer of the atmosphere being considered. The
figure below shows the definitions of the atmospheric volumes used in the WMO observation
requirements database.
While users’ requirements may change in a stepwise mode when moving along the vertical, the
quality of satellite-derived products changes with height in a smooth way, depending mainly
on the vertical gradient of the quantity, with better performance being achieved at steeper
gradients. A step change occurs when the required vertical resolution cannot be achieved
by cross-nadir scanning instruments and limb scanning becomes necessary. For the sake of
simplicity, different product performances are taken into account for the troposphere, the
stratosphere, and the total atmospheric column (where applicable). It is understood that quality
will softly degrade with increasing altitude in the troposphere, and the same in the stratosphere.
Product quality is only quoted above the height of 1 km; below 1 km the accuracy is too irregular
and difficult to estimate.
The horizontal resolution (Δx) is the convolution of several features (sampling distance, degree
of independence of the information relative to nearby samples, the point spread function, etc.).
For simplicity, it is generally agreed to refer to the sampling distance between two successive
product values carrying independent information.
80 km 0.01 hPa
48 km 1 hPa
Stratosphere
11 km 250 hPa
0 km 1 000 hPa
Atmospheric volumes defined by users. The higher stratosphere and mesosphere (HS&M) go
together. The heights and pressures are qualitative, and refer to mid-latitudes/yearly
average. The planetary boundary layer is part of the lower troposphere.
CHAPTER 5. SPACE-BASED OBSERVATION OF GEOPHYSICAL VARIABLES 119
The horizontal resolution of the geophysical variable being measured is controlled by instrument
features (primarily the IFOV), the sampling distance, or pixel, and the modulation transfer
function) and by the processing scheme that may be designed to take account of interfering
effects (such as clouds in the IFOV). For example, if clouds prevent the measurement being
useful, it may be convenient to process pixel arrays searching or extrapolating for the less
contaminated measurement in the cell of size Δx. The number of pixels to be co-processed
depends on the spectral range used to perform the measurement (down to one for all-weather
microwaves, for example) and on the available spectral information (when more spectral
channels are available, a smaller cluster of pixels is needed). The extreme case is when a large
pixel array (32 x 32, for instance) is needed to characterize the geophysical variable (an example
is the inference of atmospheric motion vectors from the displacement of highly correlated cloudy
pixel arrays within two images at different times).
For parameters such as cloud cover or snow cover, a sufficient number of samples (pixels) in
the Δx · Δx cell is necessary to achieve the required accuracy. For cloud-disturbed surface
measurements of slowly-changing variables (such as snow) it may be necessary to apply a multi-
temporal analysis that waits for the clouds to move away (this would be a Level 3 product). It
is generally possible, within limits, to trade off horizontal resolution against accuracy during
product generation. Often, the product horizontal resolution is larger than a single pixel in order
to enhance the SNR to meet the product accuracy requirements.
For cross-nadir scanning instruments, the instrument IFOV or pixel size gets larger from the
sub-satellite point towards the swath edge; the product horizontal resolution performance must
therefore be averaged across the instrument swath.
For conical scanners, the along-scan resolution is constant, but the cross-scan resolution is
degraded by the cosine of the azimuth angle (the IFOV is nearly elliptical). The quadratic average
in the along- and across-scan directions must be considered, and account has to be taken of
the IFOV elongation in the along-scan direction due to the line-of-scan motion during the
measurement integration time. If a single antenna is used for several frequencies, the resolution
will change with frequency due to diffraction.
For limb sounding, the horizontal resolution is determined by the viewing geometry. The
atmospheric path may physically extend for a few thousand kilometres, but the effective path
(which accounts for higher atmospheric density around the tangent height) is around 300 km
along-view. Across the viewing direction, although the transversal IFOV may be much narrower
(tens of kilometres), the product resolution is determined by the number of azimuth views (in
most cases only one, fore or aft). For the sake of simplicity, the typical horizontal resolution of
limb measurements is taken as 300 km.
The vertical resolution (Δz) is also defined by referring to the vertical sampling distance between
two successive product values, carrying independent information.
The vertical resolution of the product depends on the sensing principle, the instrument spectral
range and the number of channels or spectral resolution. The weighting function may be more
or less broadened in the vertical depending on the spectral resolution and range (worse in
MW, better in the optical ranges). Moreover, the spectral channels may be narrow enough to
observe single lines of the absorbing/emitting gas, or a few lines or line bands. If several lines are
included in the channel, the weighting function will be broadened since it will average surface
emission between the lines (peaking in the lower atmosphere) and atmospheric emission in
the lines (peaking at higher altitudes). In general, resolving power λ/Δλ ≈ 100 enables broad-
resolution retrieval of temperature vertical profiles with roughly 2 km vertical resolution; λ/
Δλ ≈ 1 000 enables higher vertical resolution retrieval of temperature at about 1 km along with
total-column retrieval of trace gases; λ/Δλ ≈10 000 is needed for trace-gas profiles. The gas
density has a bearing on the achievable vertical resolution, so that with increasing altitude the
vertical measurement resolution degrades, becoming unacceptable in the medium and high
stratosphere.
120 GUIDE TO INSTRUMENTS AND METHODS OF OBSERVATION - VOLUME IV
It should be noted that the weighting function shifts to higher altitudes as the instrument
viewing angle shifts from nadir to swath edge. This is due to the longer path length through
the atmosphere with increasing view angle. The transmittance is an exponential function of
the number of absorbing molecules in the path of the escaping radiation; a more oblique angle
entails a greater likelihood of encountering more molecules in the upper atmosphere and hence
the weighting function moves up in altitude.
The vertical resolution depends on the sensitivity of the wavelength to temperature. IR sensitivity
to temperature is higher in the MWIR range (about 3–6 µm), thus the weighting functions are
narrower in the lower troposphere and very broad in the higher troposphere and stratosphere.
Short waves are less sensitive to temperature, so the vertical resolution is relatively homogeneous
with altitude. MW is relatively more sensitive to cold temperature and the vertical resolution is
relatively good in the stratosphere.
In the stratosphere and above, the vertical resolution achievable by cross-nadir scanning is poor.
Limb scanning offers better vertical resolution; it is performed by mechanical scanning along the
vertical (angular IFOV combined with the scan rate) and is in the range of 1 to 3 km (which is not
possible with cross-nadir scanning). The vertical resolution achieved by limb sounding degrades
with altitude, as the SNR degrades with decreasing gas concentration. Occultation instruments
(including radio occultation) have a vertical resolution that is determined by the sampling rate
during the occultation phase. During ground processing, an algorithm performs some vertical
integration as a trade-off against product accuracy.
The observing cycle (Δt) is defined as the time required to achieve global coverage (for LEO) or
full disk coverage (for GEO). It is closely linked to the scanning capability of the instrument and
to the orbital features. The relationship between observing cycle and scanning mechanism has
been extensively discussed in the present volume, Chapter 3, 3.1.1. However, the instrument
observing cycle may not coincide with the product observing cycle since not all observations
taken during an instrument observing cycle may be useful for a given product. For example,
a clear-sky mapped product may exhibit too many gaps due to cloud-affected observations.
Thus the effective product observing cycle is a compromise between the minimum theoretical
observing cycle that will have many gaps, and multi-temporal analysis degrading the product
observing cycle but producing a more regular product field (generated by a Level 3 process).
The compromise takes into account the sensitivity of the spectral band to the disturbing factor
and the intrinsic time-variability of the desired geophysical parameter (which might not tolerate
delays entailed by multi-temporal analysis). In another example, multi-temporal analysis might
be necessary to collect enough signal when the required product accuracy cannot readily be
achieved.
For most meteorological variables the required observing cycle prevents multi-temporal analysis.
The solution is at system level, by establishing the number of satellites available to measure
the geophysical variable. A global observing cycle shorter than 12 h (for measurements in IR
and MW) or 24 h (for measurements involving SW) requires more satellites in regularly spaced
orbits. For a 3 h cycle, four satellites are needed, provided the instrument swath is as large as the
decalage (VIS/IR imagers, for example). For limited-swath instruments (such as MW radiometers
of the GPM mission) the 3 h cycle requires eight satellites.
The observing cycle may be shortened at the expense of global coverage by using low-
inclination orbits. The extreme limit is Δt < one orbital period for a quasi-equatorial orbit run
from east to west. Latitudes beyond the reach of the instrument swath will not be covered.
For satellites in GEO orbits, the observing cycle depends on the instrument refresh cycle. It may
be minutes if the observation is unaffected by clouds; otherwise multi-temporal analysis may be
needed. A constellation of six regularly-spaced GEO satellites provides coverage of all latitudes
below 55°, rising to 70° and above for longitudes close to that of the six GEO locations.
CHAPTER 5. SPACE-BASED OBSERVATION OF GEOPHYSICAL VARIABLES 121
Instruments with only nadir-viewing (non-scanning) provide infrequent global coverage. Limb-
scanning instruments, including radio occultation, have a similar drawback (see the present
volume, Chapter 3, 3.1.1). For these instruments, the observing cycle is difficult to define.
Accuracy is defined as the “closeness of the agreement between a measured quantity value and
a true quantity value of the measurand” (from the International Vocabulary of Metrology – Basic and
General Concepts and Associated Terms (VIM), Joint Committee for Guides in Metrology (JCGM)
200: 2012). The quantitative expression corresponding to the accuracy is the uncertainty (see
Volume I, Chapter 1, 1.6.2). It is the combined result of several instrument features: random
error, bias, sensitivity, precision, and the like. In the present volume (space-based observations),
the uncertainty is generally characterized by the root-mean-square (RMS) error range, namely
the RMS difference (observed – true values) of the measurement. The uncertainty of a satellite-
derived observation of a geophysical variable is driven by the physical principle linking the
satellite measurement to the observed variable, and in particular by the sensitivity of the
measurement to variations of this variable.
The radiometric resolution is thus a driving factor of the product uncertainty. It can be
characterized by the noise equivalent differential temperature, or the signal-to-noise ratio, or the
noise equivalent spectral radiance, as defined in the present volume, Chapter 3, 3.1.4. However,
the product uncertainty is also strongly affected by the retrieval algorithm and by the trade-off
with the other quality features (Δx, Δz and Δt). Furthermore, the nature of the target (intensity
of the emitted or scattered signal), the sensitivity of the sensing technique to the geophysical
variable, and the efficiency in filtering out disturbing factors (such as clouds) have a pronounced
impact on the final product uncertainty.
For a new instrument, evaluating the uncertainty requires sensitivity studies based on
complicated simulations.
In the present volume, product uncertainty is estimated from the heritage of past and current
instruments, and simulation of planned instruments. Some validation of satellite-derived product
accuracy (discussed in the present volume, Chapter 6) enters into estimates for past and current
instruments; for future instruments a theoretical calculation is performed.
5.1.2.6 Timeliness
Timeliness (δ) is defined as the time elapsed between the moment the observation is taken and
the availability of the product assuming routine operations. Timeliness depends on the satellite
transmission facilities, the availability of acquisition stations, the processing time required to
generate the product, and the overall data management.
In this Guide, the timeliness δ of the various products has not been evaluated because it is a
system feature that is not determined solely by the instrument.
This chapter provides an overview of the satellite products that can potentially be retrieved from
current or planned instruments, for geophysical variables of the following eight themes:
This list of observation products is limited to “elementary” geophysical variables; it does not
include products that can be derived from other products.
For each satellite product, the applicable remote sensing principles are indicated, together with
any observing conditions or limitations. The annex to the present chapter contains an evaluation
of the achievable quality in terms of RMS error,1 Δx, Δz and Δt, based on the characteristics
of state-of-the art instruments that are being developed at the time of writing this Guide, and
expected to be operational by 2020.
Table 5.1 lists basic variables for weather prediction, including NWP, that are observable from
space.
Other basic variables such as atmospheric pressure, temperature and humidity at surface; and
wind vertical component are not included because they cannot be reliably measured from space
with current technology.
Definition: 3D field of the atmospheric temperature – Required from surface to top of atmosphere
(TOA) (layers: LT, HT, LS, HS&M) – Physical unit: kelvin (K) – Uncertainty unit: K.
Method 3: GNSS (consisting of GPS and the Russian satellite navigation system, GLONASS) radio-
occultation – Principle: Atmospheric refraction of L-band signals from the global navigation
satellite system received by a LEO satellite during the occultation phase. Applicable only in LEO.
1
Note that RMS error used in this volume corresponds approximately to an expanded uncertainty with a coverage
factor of k = 1, while in the rest of this Guide a coverage factor of k = 2 is generally used (see Volume I, Chapter 1,
1.6.3.3 and the Evaluation of Measurement Data – Guide to the Expression of Uncertainty in Measurement (JCGM
100:2008)).
CHAPTER 5. SPACE-BASED OBSERVATION OF GEOPHYSICAL VARIABLES 123
Method 4: Limb sounding – Principle: Emission by lines (in IR or MW) or line broadening (in SW)
as observed by high-resolution spectrometers intended for atmospheric chemistry operating in
the Earth's limb. Applicable only in LEO.
Definition: 3D field of the specific humidity in the atmosphere – Required from surface to TOA
(layers: LT, HT, LS, HS&M) + total column – Physical units: g/kg for layers, kg/m2 for total column
– Uncertainty unit: % for layers, kg/m2 for total column.
Method 2: MW/sub-mm radiometry – Principle: MW and sub-mm wave emission from different
atmospheric layers, selected by using spectral intervals of different absorption strength in bands
of H2O (183 and potentially 324, 380 GHz and others at higher frequencies) with necessary
support of O2 (~54, 118 and potentially 425 GHz). Applicable in LEO and potentially in GEO.
Method 3: GNSS radio-occultation – Principle: Atmospheric refraction of L-band signals from the
GNSS received by a LEO satellite during the occultation phase. Applicable only in LEO.
Method 5: Limb sounding – Principle: Emission (in IR or MW/sub-mm), absorption (in Sun or star
occultation of SW) or scattering (in SW) by lines as observed by high-resolution spectrometers
intended for atmospheric chemistry operating in the Earth's limb. Applicable only in LEO.
Method 6: IR split window – Principle: By-product of the retrieval of surface temperatures from IR
images (split-windows such as 11 and 12 µm). Suitable only for total column. Applicable in both
LEO and GEO.
Method 8: NIR imaging (935 nm) – Principle: Differential reflectance between an NIR water-
vapour absorption band (~935 nm for example) and a side window by narrow-bandwidth
radiometers. Suitable only for total column. Applicable in both LEO and GEO.
Definition: 3D field of the horizontal vector component (2D) of the 3D wind vector – Required
from surface to TOA (layers: LT, HT, LS, HS&M) – Physical unit: m/s – Uncertainty unit: m/s
intended as vector error, in other words the module of the vector difference between the
observed vector and the true vector.
Method 2: VIS/IR image sequences – Principle: Motion of atmospheric cells of determined size
"signed" by clouds and water-vapour patches (and possibly ozone patches) recognized and
tracked in VIS/IR image sequences. Applicable in both LEO and GEO.
Method 4: Limb sounding – Principle: Doppler shift and broadening of spectral lines of O2, O3,
OH– observed by high-resolution VIS spectrometers operating in the Earth's limb. Applicable
only in LEO.
Definition: Horizontal vector component (2D) of the 3D wind vector, conventionally measured at
10 m height – Required over sea and land surface (all methods below apply over sea). Physical
unit: m/s – Uncertainty unit: m/s intended as vector error, namely the module of the vector
difference between the observed vector and the true vector.
Method 4: Radar altimetry – Principle: Backscattered radiation from sea surface by medium-
frequency radar (~13 GHz). Wind speed associated with echoes scattered from capillary waves.
Applicable only over the sea. Only speed measured. Only nadir. Applicable only in LEO.
Definition: Height of the surface separating the planetary boundary layer from the free
atmosphere – Physical unit: km – Uncertainty unit: km.
Method 1: Backscatter lidar – Principle: Backscattered radiation in UV, VIS or NIR by lidar.
Applicable only in LEO.
Method 2: From IR sounding – Principle: Derived from IR sounding of temperature and humidity.
Applicable in both LEO and GEO.
Method 3: From GNSS sounding – Principle: Derived from GNSS radio-occultation sounding of
temperature and humidity. Applicable only in LEO.
Definition: Height of the surface separating the troposphere from the stratosphere – Physical
unit: km – Uncertainty unit: km.
Method 1: Backscatter lidar – Principle: Backscattered radiation in UV, VIS or NIR by lidar.
Two wavelengths preferred. Applicable only in LEO.
Method 3: From GNSS sounding – Principle: Derived from GNSS radio-occultation sounding of
temperature. Applicable only in LEO.
CHAPTER 5. SPACE-BASED OBSERVATION OF GEOPHYSICAL VARIABLES 125
Definition: Atmospheric temperature at the height of the surface separating the troposphere from
the stratosphere – Physical unit: K – Uncertainty unit: K.
Method 2: From GNSS sounding – Principle: Derived from GNSS radio-occultation sounding of
temperature. Applicable only in LEO.
Method 3: From limb sounding – Principle: Derived from limb sounding of temperature.
Applicable only in LEO.
This theme includes the basic variables observable from space for actual weather analysis, and for
short-term prediction and nowcasting, as well as for hydrology. Table 5.2 lists these variables.
Cloud-top height Cloud optical depth Cloud-ice effective radius Precipitation intensity at
surface (liquid or solid)
Definition: Height of the upper surface of the cloud – Physical unit: km – Uncertainty unit: km.
Method 1: IR radiometry – Principle: Derived from cloud-top pressure converted to height and
temperature using a forecast temperature profile after viewing the cloud through an 11 and
13.4 µm channel pair (comparison with “window”, measures the columnar defect of CO2 above
cloud top). Applicable in both LEO and GEO.
126 GUIDE TO INSTRUMENTS AND METHODS OF OBSERVATION - VOLUME IV
Method 3: Backscatter lidar – Principle: Backscattered radiation in UV, VIS or NIR by lidar. Two
wavelengths preferred. Applicable only in LEO.
Method 5: A-band spectroscopy – Principle: Observed defect of columnar O2 above the cloud top
by spectroscopy of the 760 nm A-band and nearby “window”. Applicable in both LEO and GEO.
Definition: Comprehensive properties of the observed cloud. The list of types of interest is
predetermined – Uncertainty expressed as number of discriminated types (classes).
Definition: 3D field of fraction of sky where clouds are detected. Required as a 3D field in the
troposphere (assumed height: 12 km) and also as a single layer (total column) to provide the total
cloud cover – Physical unit: % – Uncertainty unit: %.
Method 1: VIS/IR radiometry – Principle: Derived from cloud imagery in a few discrete channels
selected so as to detect all cloud types. The fractional cover refers to the number of cloudy pixels
in a given pixel array. Applicable in both LEO and GEO.
Method 3: Cloud radar – Principle: Backscattered radiation from cloud droplets observed by high-
frequency radar (~94 GHz). Applicable only in LEO.
Definition: Height of the bottom surface of the cloud – Physical unit: km – Uncertainty unit: km.
Method 1: Cloud radar – Principle: Derived as lower level of the backscattered radiation from
cloud droplets observed by high-frequency radar (~94 GHz). Applicable only in LEO.
Definition: Effective depth of the cloud from the viewpoint of radiation propagation. The
definition is OD = e–K Δz. K is the extinction coefficient (km–1), Δz the optical path (km) between
the base and the top of the cloud. It depends on the wavelength but is usually referred to visible
radiation. Physical unit: dimensionless – Uncertainty unit: dimensionless.
CHAPTER 5. SPACE-BASED OBSERVATION OF GEOPHYSICAL VARIABLES 127
Method 1: Backscatter lidar – Principle: Backscattered radiation in UV, VIS or NIR by lidar.
Two wavelengths preferred. Applicable only in LEO.
Definition: 3D field of atmospheric water in the liquid phase (precipitating or not). Required in the
troposphere (assumed height: 12 km) and for total column – Physical unit: g/kg for layers,
g/m2 for total column – Uncertainty unit: % for layers, g/m2 for total column.
Method 1: Cloud radar – Principle: Backscattered radiation from cloud droplets observed by high-
frequency radar (~94 GHz). Applicable only in LEO.
Method 2: Precipitation radar – Principle: Backscattered radiation from cloud droplets observed
by medium-frequency radar (dual-frequency preferred, 14 and 35 GHz). Applicable only in LEO.
Definition: 3D field of the size distribution of liquid water droplets, assimilated to spheres of the
same volume. Required in the troposphere (assumed height: 12 km), and at the cloud-top surface
– Physical unit: µm – Uncertainty unit: µm.
Method 1: Cloud radar – Principle: Backscattered radiation from cloud droplets by high-frequency
radar (~94 GHz). Applicable only in LEO.
Method 4: Backscatter lidar – Principle: Backscattered radiation in UV, VIS or NIR by lidar.
Essentially limited to cloud top. Applicable only in LEO.
Method 6: VIS/IR radiometry – Principle: Scattered solar radiation in several narrowband channels
of VIS, NIR, SWIR and MWIR. Also, differential emission in several channels of thermal infrared
(TIR, for cirrus clouds). Essentially limited to cloud top. Applicable in both LEO and GEO.
128 GUIDE TO INSTRUMENTS AND METHODS OF OBSERVATION - VOLUME IV
Definition: 3D field of atmospheric water in the solid phase (precipitating or not). Required in the
troposphere (assumed height: 12 km) and for total column – Physical unit: g/kg for layers,
g/m2 for total column – Uncertainty unit: % for layers, g/m2 for total column.
Method 1: Cloud radar – Principle: Backscattered radiation from ice particles observed by high-
frequency radar (~94 GHz). Applicable only in LEO.
Method 2: Precipitation radar – Principle: Backscattered radiation from ice particles by medium-
frequency radar (dual-frequency preferred, 14 and 35 GHz). Applicable only in LEO.
Method 5: FIR imagery – Principle: Emitted and scattered radiation in several atmospheric
windows of FIR (18.2, 24.4, 52, 87 µm) as compared to TIR (8.7, 11, 12 µm). Suitable only for total
column. Applicable only in LEO.
Definition: 3D field of the size distribution of ice particles, assimilated to spheres of the same
volume. Required in the troposphere (assumed height: 12 km), and at the cloud-top surface –
Physical unit: µm – Uncertainty unit: µm.
Method 1: Cloud radar – Principle: Backscattered radiation from ice particles by high-frequency
radar (~94 GHz). Applicable only in LEO.
Method 2: Precipitation radar – Principle: Backscattered radiation from ice particles by medium-
frequency radar (dual-frequency preferred, 14 and 35 GHz). Applicable only in LEO.
Method 4: Backscatter lidar – Principle: Backscattered radiation in UV, VIS or NIR by lidar.
Essentially limited to cloud top. Applicable only in LEO.
Method 6: VIS/IR radiometry – Principle: Scattered solar radiation in several narrowband channels
of VIS, NIR, SWIR and MWIR. Also, differential emission in several channels of TIR (for cirrus
clouds). Essentially limited to cloud top. Applicable in both LEO and GEO.
Definition: Height of the atmospheric layer in cloud where liquid-solid states transform into each
other – Physical unit: km – Uncertainty unit: km.
CHAPTER 5. SPACE-BASED OBSERVATION OF GEOPHYSICAL VARIABLES 129
Method 1: Precipitation radar – Principle: Backscattered radiation from cloud drops by medium-
frequency radar (dual-frequency preferred, 14 and 35 GHz). Applicable only in LEO.
Method 2: From MW/sub-mm sounding – Principle: Derived from MW and sub-mm wave
sounding of temperature. Applicable in LEO and potentially in GEO.
Definition: Depth of the atmospheric layer in cloud where liquid-solid states transform into each
other – Physical unit: km – Uncertainty unit: km.
Method 1: Precipitation radar – Principle: Backscattered radiation from cloud drops by medium-
frequency radar (dual-frequency preferred, 14 and 35 GHz). Applicable only in LEO.
Method 2: From MW/sub-mm sounding – Principle: Derived from MW and sub-mm wave
sounding of temperature. Applicable in LEO and potentially in GEO.
Definition: 3D field of the vertical flux of precipitating water mass. Required in the troposphere
(assumed height: 12 km) – Physical unit: g · s–1 · m–2 (vertical flux of precipitating water mass) –
Uncertainty unit: %.
Method 1: Precipitation radar – Principle: Backscattered radiation from cloud drops by medium-
frequency radar (dual-frequency preferred, 14 and 35 GHz). Doppler capability also useful.
Applicable only in LEO.
Definition: Intensity of precipitation reaching the ground – Physical unit: mm/h (if solid, mm/h of
liquid water after melting) – Uncertainty unit: mm/h. Since uncertainty changes with intensity, it
is necessary to specify a reference intensity. Assumed intensity: 5 mm/h.
Method 1: Precipitation radar – Principle: Backscattered radiation from cloud drops by medium-
frequency radar (dual-frequency preferred, 14 and 35 GHz). Doppler capability also useful.
Applicable only in LEO.
Method 3: VIS/IR radiometry – Principle: Inferred from cloud imagery in a few discrete channels
selected so as to detect all cloud types, assisted by conceptual models, generally more responsive
to convective rain. Applicable in GEO.
Method 4: Fusion between MW from LEO and IR from GEO – Principle: Combined product of
LEO/MW-derived accurate/infrequent measurements with GEO/IR frequent images used either
to be “calibrated” by MW measurements or to enable dynamical interpolation between MW-
derived precipitation data. Requiring both LEO and GEO.
130 GUIDE TO INSTRUMENTS AND METHODS OF OBSERVATION - VOLUME IV
Definition: Integration of precipitation intensity reaching the ground in given time intervals. The
reference requirement refers to integration over 24 h – Physical unit: mm – Uncertainty unit: mm.
Method 1: From fusion between MW from LEO and IR from GEO – Principle: Derived by time
integration of frequent precipitation rate measured by merging MW precipitation rate data from
LEO with IR imagery from GEO. Requiring both LEO and GEO.
Definition: Mapping of lightning events as number of flashes in a given time interval over a given
area – Physical unit: counts – Uncertainty expressed as hit rate (HR) and false-alarm rate (FAR).
This theme comprises variables that affect the Earth radiation budget versus space, cloud–
radiation interaction, cloud formation, air quality and several characterizing factors of climate
and climate change. The variables observable from space are listed in Table 5.3.
Table 5.3. Geophysical variables considered under the theme “Aerosol and radiation”
Aerosol optical depth Upward spectral radiance at TOA Earth’s surface albedo
Aerosol effective radius Upward SW irradiance at TOA Upward LW irradiance at Earth’s surface
Further variables such as aerosol absorption optical depth, aerosol extinction coefficient, aerosol
single scattering albedo and aerosol phase function have not been considered since they are
closely linked to the selected ones (optical thickness, concentration, effective radius, and type),
which are more understandable to the general user.
Definition: Effective depth of the aerosol column from the viewpoint of radiation propagation.
The definition is OD = exp (–K Δz). K is the extinction coefficient (km–1), Δz the optical
path (km) between the Earth’s surface and TOA. It depends on the wavelength – Physical
unit: dimensionless – Uncertainty unit: dimensionless.
CHAPTER 5. SPACE-BASED OBSERVATION OF GEOPHYSICAL VARIABLES 131
Method 1: Backscatter lidar – Principle: Backscattered radiation in UV, VIS or NIR by lidar.
Applicable only in LEO.
Method 3: SW spectroscopy (cross-nadir) – Principle: Scattered radiation in UV, VIS, NIR and SWIR
observed cross-nadir with high spectral resolution. Applicable in both LEO and GEO.
Method 4: VIS/IR radiometry – Principle: Scattered solar radiation in several channels of VIS, NIR
and SWIR. Some information, relevant to absorbing aerosol, also available in thermal IR windows.
Applicable in both LEO and GEO.
Definition: 3D field of the mass mixing ratio of condensed particles in the atmosphere (other than
water) – Required from surface to TOA (layers: LT, HT, LS, HS&M) + total column – Physical units:
g/kg for layers, g/m2 for total column – Uncertainty unit: % for layers, g/m2 for total column.
Method 1: Backscatter lidar – Principle: Backscattered radiation in UV, VIS or NIR by lidar.
Applicable only in LEO.
Method 3: SW spectroscopy (cross-nadir) – Principle: Scattered radiation in UV, VIS, NIR and SWIR
observed cross-nadir with high spectral resolution. Applicable in both LEO and GEO.
Method 4: SW spectroscopy (limb) – Principle: Scattered radiation in UV, VIS, NIR and SWIR
observed in limb mode with high spectral resolution. Also, absorbed radiation from Sun or stars
during occultation. Applicable only in LEO.
Method 5: VIS/IR radiometry – Principle: Scattered solar radiation in several channels of VIS, NIR
and SWIR. Some information, relevant to absorbing aerosol, also available in thermal IR windows.
Suitable only for total column. Applicable in both LEO and GEO.
Definition: 3D field of the mean size of the aerosol particles, assimilated to spheres of the same
volume. Required in the troposphere (assumed height: 12 km) and as columnar average –
Physical unit: µm – Uncertainty unit: µm.
Method 1: Backscatter lidar – Principle: Backscattered radiation in UV, VIS or NIR by lidar.
Applicable only in LEO.
Method 3: SW spectroscopy (cross-nadir) – Principle: Scattered radiation in UV, VIS, NIR and SWIR
observed cross-nadir with high spectral resolution. A priori information and intensive modelling
necessary. Applicable in both LEO and GEO.
132 GUIDE TO INSTRUMENTS AND METHODS OF OBSERVATION - VOLUME IV
Definition: 3D field. Comprehensive properties of the aerosol being observed. The list of types
of interest is predetermined – Required in the troposphere (assumed height: 12 km) and as
columnar average – Uncertainty expressed as number of types that can actually be resolved
(classes).
Method 1: Backscatter lidar – Principle: Backscattered radiation in UV, VIS or NIR by lidar.
Applicable only in LEO.
Method 3: SW spectroscopy (cross-nadir) – Principle: Scattered radiation in UV, VIS, NIR and SWIR
observed cross-nadir with high spectral resolution. A priori information and intensive aerosol
modelling absolutely necessary. Applicable in both LEO and GEO.
Method 4: VIS/IR radiometry – Principle: Scattered solar radiation in several channels of VIS, NIR
and SWIR. Some information also available in thermal IR windows. A priori information and
intensive aerosol modelling necessary. Suitable only for total column. Applicable in both LEO
and GEO.
Definition: 3D field of concentration of volcanic ash – Required from surface to TOA (layers: LT, HT,
LS, HS&M) + total column – Physical units: g/kg for layers, g/m2 for total column – Uncertainty
unit: % for layers, g/m2 for total column.
Method 1: Backscatter lidar – Principle: Backscattered radiation in UV, VIS or NIR by lidar.
Applicable only in LEO.
Method 3: SW spectroscopy (cross-nadir) – Principle: Scattered radiation in UV, VIS, NIR and SWIR
observed cross-nadir with high spectral resolution. Applicable in both LEO and GEO.
Method 4: SW spectroscopy (limb) – Principle: Scattered radiation in UV, VIS, NIR and SWIR
observed in limb mode with high spectral resolution. Also, absorbed radiation from Sun or stars
during occultation. Applicable only in LEO.
Method 5: VIS/IR radiometry – Principle: Scattered solar radiation in several channels of VIS, NIR
and SWIR. Some information also available in thermal IR windows. Suitable only for total column.
Applicable in both LEO and GEO.
Definition: Flux density of the solar radiation at the top of the atmosphere – Physical unit: W/m2 –
Uncertainty unit: W/m2.
Method 1: Cavity radiometer – Principle: Trapping of total downward solar radiation at satellite
altitude into devices such as active cavities. Absolute measurement. Applicable in LEO, GEO, or
outer-space orbits, at the L1 Lagrange libration point, for instance.
CHAPTER 5. SPACE-BASED OBSERVATION OF GEOPHYSICAL VARIABLES 133
Definition: Flux density of terrestrial radiation emitted to space by the Earth’s surface, atmosphere
and clouds at the top of the atmosphere – Physical unit: W/m2 – Uncertainty unit: W/m2.
Definition: Flux density of terrestrial radiation reflected to space by the Earth’s surface,
atmosphere and clouds at the top of the atmosphere – Physical unit: W/m2 – Uncertainty unit: W/
m2 .
Definition: Reflectance of the solar radiation from clouds – Physical unit: % – Uncertainty unit: %.
Method 1: SW radiometry – Principle: Scattered solar radiation in several channels of VIS, NIR and
SWIR. Multi-viewing geometry useful. Applicable in both LEO and GEO.
Definition: Flux density of LW radiation from Sun, atmosphere and clouds to the Earth’s surface –
Physical unit: W/m2 – Uncertainty unit: W/m2.
Method 1: From IR/MW sounding – Principle: High-level product derived mostly from
atmospheric temperature and water-vapour profiles. Contributions also from cloud cover
profile, specifically cloud-base height. Atmospheric modelling necessary. Applicable in both LEO
and GEO.
Definition: Flux density of SW radiation from Sun, atmosphere and clouds to the Earth’s surface –
Physical unit: W/m2 – Uncertainty unit: W/m2.
134 GUIDE TO INSTRUMENTS AND METHODS OF OBSERVATION - VOLUME IV
Definition: Hemispherically integrated reflectance of the Earth’s surface in the range 0.4–0.7 µm
(or other specific SW ranges) – Physical unit: % – Uncertainty unit: %.
Definition: Reflectance of the Earth’s surface as a function of the viewing angle and the
illumination conditions in the range 0.4–0.7 µm (or other specific SW ranges) – Physical unit: % –
Uncertainty unit: %.
Method 1: SW radiometry – Principle: Scattered solar radiation in several channels of VIS, NIR
and SWIR observed under several viewing angles and solar angles to estimate anisotropy effects
and improve radiative flux computations. Channels for atmospheric corrections also included.
Applicable only in LEO.
Definition: Flux density of long-wave radiation emerging from the Earth’s surface – Physical
unit: W/m2 – Uncertainty unit: W/m2.
Definition: Emissivity of the Earth’s surface in the thermal IR, function of the wavelength – Physical
unit: % – Uncertainty unit: %.
Method 1: VIS radiometry – Principle: Computed from the observed photosynthetically active
radiation and one measure in the “red” region (~670 nm). Applicable in both LEO and GEO.
This theme comprises variables that characterize the ocean surface, including waves and sea ice.
The variables observable from space are listed in Table 5.4.
Table 5.4. Geophysical variables considered under the theme “Ocean and sea ice”
Ocean chlorophyll Oil-spill cover Coastal sea level (tide) Wave directional-energy
concentration frequency spectrum
Colour dissolved organic Sea-surface temperature Significant wave height Sea-ice cover
matter
Ocean diffuse attenuation Ocean dynamic Dominant wave period Sea-ice type
coefficient topography
Many variables have not been considered: underwater profiles of temperature and salinity
(impossible to measure from space), currents (derivable from ocean topography as for the
geostrophic component, otherwise impossible or too inaccurate), iceberg extension or height
(special case of ice cover and thickness), and ice drift (product of multi-temporal analysis).
Method 1: VIS radiometry – Principle: Measurement of reflected solar radiation in several channels
(most significant: 442.5 nm, 490 nm, 560 nm, 665 nm, 681.25 nm). Spectral resolution of the
order of 2%. Applicable in both LEO and GEO.
136 GUIDE TO INSTRUMENTS AND METHODS OF OBSERVATION - VOLUME IV
Definition: Former name: “Yellow substance absorbance”. Variable extracted from ocean-colour
observation. Indicative of biomass undergoing decomposition processes. Required in both open
ocean and coastal zone – Physical unit: m–1 – Uncertainty unit: m–1 at a specific concentration
(such as 1 m–1).
Method 1: VIS radiometry – Principle: Measurement of reflected solar radiation in several channels
(most significant: 412.5 nm). Spectral resolution of the order of 2%. Applicable in both LEO
and GEO.
Definition: Variable extracted from ocean-colour observation. Indicative of river outflow, re-
suspension or pollution of other-than-biological origin. Required in both open ocean and coastal
zone – Physical unit: g/m3 – Uncertainty unit: g/m3 at a specific concentration
(such as 2 g/m3).
Method 1: VIS radiometry – Principle: Measurement of reflected solar radiation in several channels
(most significant: 510 nm, 560 nm, 620 nm). Spectral resolution of the order of 2%. Applicable in
both LEO and GEO.
Definition: Former name: "Water clarity". Indicator of water turbidity and vertical processes in the
ocean, extracted from ocean-colour observation. Required in both open ocean and coastal zone
– Physical unit: m–1 – Uncertainty unit: m–1.
Method 1: VIS radiometry – Principle: Measurement of reflected solar radiation in several channels
of the range 400–700 nm. Spectral resolution of the order of 2%. Applicable in both LEO
and GEO.
Definition: The fraction of an ocean area polluted by hydrocarbons released from ships,
accidentally or deliberately. Oil spills are impacting on ocean-atmosphere exchanges. Required
in both open ocean and coastal zone – Physical unit: % – Uncertainty unit: %.
Definition: Temperature of the sea water at surface. The “bulk” temperature refers to the depth of
typically 2 m, the “skin” temperature refers to values within the upper 1 mm. – Physical unit: K –
Uncertainty unit: K.
Method 2: IR spectroscopy – Principle: Derived from a high number of very narrow channels
through the IR spectrum, associated with other channels providing all the information needed
for atmospheric corrections. Applicable in both LEO and GEO.
Definition: Salinity of sea water in the surface layer, which is the layer affected by turbulence
associated with wind stress, waves and diurnal solar heating cycle. (The layer is a few metres
deep but a microwave observation would be representative of the upper ~1 m). In the open
ocean the correct term should be “halinity” in order to make reference to the most common
anion, chlorine – Physical unit: practical salinity unit (PSU), close to 1‰, or 1 g of salt per 1 litre of
solution – Uncertainty unit: PSU.
Definition: Deviation of sea level from the geoid caused by ocean currents (after corrections for
tides and atmospheric pressure effects) – Physical unit: cm – Uncertainty unit: cm.
Method 1: Radar altimetry – Principle: Backscattered radiation from sea surface by medium-
frequency radar (dual-frequency preferred, 13 and 3 or 5 GHz). Associated with two or three
channels of passive MW radiometry (23 and 37 and/or 19 GHz) needed for tropospheric path
correction from water vapour and ionosphere-induced rotation. Ocean topography is extracted
by filtering the fluctuation of wave heights out of the satellite-to-surface measured range.
Applicable only in LEO.
Definition: Deviation of sea level from local references in coastal zones, caused by local currents
and tides (astronomical and wind-induced) – Physical unit: cm – Uncertainty unit: cm.
Method 1: Radar altimetry – Principle: Backscattered radiation from sea surface by medium-
frequency radar (dual-frequency preferred, 13 and 3 or 5 GHz). Associated with two or three
channels of passive MW radiometry (23 and 37 and/or 19 GHz) needed for tropospheric path
correction from water vapour and ionosphere-induced rotation. Sea level is extracted by filtering
the fluctuation of wave heights out of the satellite-to-surface measured range. Applicable only
in LEO.
138 GUIDE TO INSTRUMENTS AND METHODS OF OBSERVATION - VOLUME IV
Definition: Average amplitude of the highest 30 of 100 waves – Physical unit: m – Uncertainty unit:
m.
Method 1: Radar altimetry – Principle: Backscattered radiation from sea surface by medium-
frequency radar (dual-frequency preferred, 13 and 3 or 5 GHz). Wave height is linked to the
statistical dispersion of the radar-measured ranges. Applicable only in LEO.
Method 2: From SAR spectra – Principle: From spectral analysis of SAR images at frequencies of 1.3
or 5 GHz by processing spectrum power, wavelength and direction with the help of boundary
conditions. Applicable only in LEO.
Definition: One feature of the ocean wave spectrum. It is the direction of the most energetic wave
in the spectrum – Physical unit: degrees – Uncertainty unit: degrees.
Method 1: From SAR spectra – Principle: From spectral analysis of SAR images at frequencies of
1.3, 5 or 11 GHz. Applicable only in LEO.
Definition: One feature of the ocean wave spectrum. It is the period of the most energetic wave in
the spectrum – Physical unit: s – Uncertainty unit: s.
Method 1: From SAR spectra – Principle: From spectral analysis of SAR images at frequencies of
1.3, 5 or 11 GHz. Applicable only in LEO.
Method 1: From SAR spectra – Principle: From spectral analysis of SAR images at frequencies of
1.3, 5 or 11 GHz. Applicable only in LEO.
Definition: The fraction of an ocean area where ice is detected – Physical unit: % – Uncertainty
unit: %.
Definition: Thickness of the ice sheet. It is related to sea-ice elevation and ice density – Physical
unit: cm – Uncertainty unit: cm.
Method 1: Radar altimetry – Principle: Backscattered radiation from sea surface by medium-
frequency radar (dual-frequency preferred, 13 and 3 or 5 GHz). Associated with two-channel
passive MW radiometry (23 and 37 GHz) needed for tropospheric path correction from water
vapour. Applicable only in LEO.
Method 2: Lidar altimetry – Principle: Backscattered VIS/NIR radiation by lidar. Two wavelengths
preferred, such as 532 and 1 064 nm. Applicable only in LEO.
Definition: Comprehensive properties (age, roughness, density, etc.) of the observed sea ice. The
list of types of interest is predetermined – Uncertainty expressed as number of discriminated
types (classes).
This theme comprises variables that characterize the land surface, including vegetation, fire,
glaciers and snow. The variables observable from space are listed in Table 5.5.
A few variables have not been considered, such as groundwater (considered covered by soil
moisture, snow, glaciers; and land cover); river discharge (products are of too high a level);
subsoil temperature profile (impossible from space); snow and lake surface temperature;
permafrost (specific cases of surface temperature observation); coastlines (too obvious); biomass
(too generic).
140 GUIDE TO INSTRUMENTS AND METHODS OF OBSERVATION - VOLUME IV
Land surface temperature Leaf area index Snow status (wet/dry) Land surface topography
Soil moisture Fire fractional cover Snow water equivalent Glacier topography
(in the roots region)
Definition: Temperature of the apparent surface of land (bare soil or vegetation) – Physical unit: K
– Uncertainty unit: K.
Method 2: IR spectroscopy – Principle: Derived from a high number of very narrow window
channels through the IR spectrum, associated with other channels providing all the information
needed for atmospheric corrections. This enables emissivity to be estimated. Applicable in both
LEO and GEO.
Definition: Fractional content of water in a volume of wet soil. Surface layer (upper few
centimetres) – Physical unit: m3/m3 – Uncertainty unit: m3/m3.
Method 1: MW radiometry – Principle: Emitted MW radiation at low frequencies (1.4 and 2.7 GHz,
for example). Multiple polarizations needed, to correct for roughness effects. Multiple channels
desirable, to correct for temperature. Higher frequencies (5, 10 GHz) also useful, particularly for
bare soil. Applicable only in LEO.
Method 4: VIS/IR radiometry – Principle: Several proxies are possible. Examples: damping of
reflectivity from VIS/NIR to SWIR; from apparent thermal inertia derived by measuring the delay
in land temperature rising in response to incoming solar radiation (valid for bare soil). Applicable
in both LEO and GEO.
CHAPTER 5. SPACE-BASED OBSERVATION OF GEOPHYSICAL VARIABLES 141
Definition: Subsoil 3D field of the fractional content of water in a volume of wet soil. Required
from surface down to ~3 m – Physical unit: m3/m3 – Uncertainty unit: m3/m3.
Definition: The fraction of a land area where vegetation is present – Physical unit: % – Uncertainty
unit: %.
Definition: Observed vegetal species or families. The list of types of interest is predetermined –
Uncertainty is expressed as number of identified types (classes).
Definition: One half of the total projected green leaf fractional area in the plant canopy within
a given area. Representative of total biomass and health of vegetation – Physical unit: % –
Uncertainty unit: %.
Method 1: SW radiometry – Principle: Scattered solar radiation through VIS/NIR and deeply
into SWIR (up to 2.4 µm, for example). Several channels needed, relatively narrow (2%–3%).
Applicable in both LEO and GEO.
Definition: Difference between maximum (in NIR) and minimum (around the “red”) vegetation
reflectance, normalized to the summation. Representative of total biomass, supportive for
computing leaf area index if not directly measured – Physical unit: % – Uncertainty unit: %.
Method 1: VIS/NIR radiometry – Principle: Scattered solar radiation in VIS (“red”, minimum
reflectance from vegetation) and NIR (typically, 865 nm, high reflectance). Applicable in both
LEO and GEO.
Method 2: High-resolution optical imagery – Principle: Scattered solar radiation in VIS (“red”,
minimum reflectance from vegetation) and NIR (typically, 865 nm, high reflectance). Applicable
in LEO and potentially in GEO.
Definition: The fraction of a land area where fire is present – Physical unit: % – Uncertainty unit: %.
Definition: Temperature of the fire occurring within an area – Physical unit: K – Uncertainty unit: K.
Definition: Power radiated by the fire occurring within an area – Physical unit: kW · m–2 –
Uncertainty unit: kW · m–2.
Definition: Binary product (dry or melting/thawing) expressing the presence of liquid water in a
snow layer – Uncertainty expressed as HR and FAR when classifying the status as either wet or
dry.
Definition: The fraction of a given area which is covered by snow – Physical unit: % – Uncertainty
unit: %.
Definition: Vertical depth of the water that would be obtained by melting a snow layer. The snow
depth may be inferred by exploiting auxiliary information on the density of the snow layer –
Physical unit: mm – Uncertainty unit: mm.
Definition: Observed soil composition or structure (acid, alkaline, rough, and the like). The list of
types of interest is predetermined – Uncertainty is expressed as number of discriminated types
(classes).
Definition: Observed land utilization (urban, cultivated, desertic, etc.). The list of types of interest
is predetermined – Uncertainty expressed as the number of identified types (classes).
Method 1: High-resolution VIS stereoscopy – Principle: Reflected solar radiation in VIS observed
in one or more channels of relatively narrow bandwidths (1%–5%) from at least two viewing
directions, generally from successive orbits, so as to implement stereoscopy. Applicable only
in LEO.
Method 3: Radar altimetry – Principle: Backscattered radiation from land surface by medium-
frequency radar (dual-frequency preferred, 13 and 3 or 5 GHz). Associated with two or three
channels of passive MW radiometry (23 and 37 and/or 19 GHz) needed for tropospheric path
correction from water vapour and ionosphere-induced rotation. Along-track SAR processing is
needed for acceptable resolution. Only nadir view. Applicable only in LEO.
Method 4: Lidar altimetry – Principle: Backscattered VIS/NIR radiation by lidar. Two wavelengths
preferred, such as 532 and 1 064 nm. Only nadir view. Applicable only in LEO.
Definition: The fraction of a land area covered by permanent ice – Physical unit: % – Uncertainty
unit: %.
Definition: Map of the height of the glacier surface – Physical unit: cm – Uncertainty unit: cm.
This theme comprises variables that characterize the solid Earth (space geodesy and Earth
interior). The variables observable from space are listed in Table 5.6.
Table 5.6. Geophysical variables considered under the theme “Solid Earth”
5.7.1 Geoid
Definition: Equipotential surface which would coincide exactly with the mean ocean surface of
the Earth, if the oceans were in equilibrium, at rest, and extended through the continents (such
as with very narrow channels) – Physical unit: cm – Uncertainty unit: cm.
Method 1: Radar altimetry – Principle: Backscattered radiation from sea surface by medium-
frequency radar (dual-frequency preferred, 13 and 3 or 5 GHz). Associated with two or three
channels of passive MW radiometry (23 and 37 and/or 19 GHz) needed for tropospheric path
correction from water vapour and ionosphere-induced rotation. Highly stable orbits needed
(relatively high altitude, 50°–70° inclination and accurate repeat cycle). Multi-orbital analysis
enables transient perturbations to be filtered out of waves, ocean currents and tides. Applicable
only in LEO.
Method 2: Gravity-field observation – Principle: Observation of the gravity field at satellite altitude
by accelerometers, gradiometers, satellite–satellite tracking (coupled satellites or with GPS
satellites). Low orbits are used, changing during mission time. Applicable only in LEO.
Definition: Basis for monitoring the evolution of the lithosphere dynamics – Physical unit: cm –
Uncertainty unit: cm.
Method 2: GPS receiver – Principle: Statistical analysis of the position of a surface-based GPS
receiver localized by the constellations of navigation satellites (GPS, GLONASS, Compass,
Galileo). Applicable only in LEO.
Definition: Changes over time of the position and height of the Earth’s plates. Indicative of the
lithosphere dynamics, thus useful for earthquake prediction – Physical unit: mm/y – Uncertainty
unit: mm/y.
Method 1: Laser ranging – Principle: Analysis of changes in crustal plate positioning, accurately
measured by satellite-ground distance through a surface-based laser that collects the light
reflected by cube-corner mirrors covering the surface of the satellite. A worldwide network of
laser-ranging stations enables this analysis to be performed. Applicable only in LEO.
Method 2: GPS receiver – Principle: Analysis of changes in crustal plate positioning, accurately
measured by surface-based GPS receivers localized by the constellations of navigation satellites
(GPS, GLONASS, Compass, Galileo). Applicable only in LEO.
146 GUIDE TO INSTRUMENTS AND METHODS OF OBSERVATION - VOLUME IV
Definition: 3D field, actually measured in situ at orbital height. Indicative of the statics and
dynamics of the lithosphere and the mantle – Physical unit: mGal (1 Gal = 0.01 m/s2,
so 1 mGal ≈ 10 –6 g0 . “Gal” stands for Galileo) – Uncertainty unit: mGal.
Definition: 3D field, actually measured in situ at orbital height. Indicative of fine details of the
statics and dynamics of the lithosphere and the mantle – Physical unit: E, Eötvös
(1 E = 1 mGal/10 km) – Uncertainty unit: E.
This theme deals with species that impact the ozone cycle, and/or provoke the greenhouse effect
and/or affect air quality. The species observable from space and which are so far the subject of
explicit requirements are listed in Table 5.7.
Table 5.7. Geophysical variables considered under the theme “Atmospheric chemistry”
Peroxyacetyl
O3 C2H2 CFC–11 CH2O ClO CO COS HCl HNO3 N2O5 NO2 SF6
nitrate
Polar stratospheric
BrO C2H6 CFC–12 CH4 ClONO2 CO2 H2O HDO N2O NO OH SO2
cloud occurrence
Definition: 3D field of dry-air mole fraction of O3. Required from surface to TOA (layers: LT, HT, LS,
HS&M) + total column – Physical unit: nmol/mol for layers, Dobson unit (DU) for total column
(1 DU = 2.69 · 1020 molecules/m2) – Uncertainty unit: nmol/mol for layers, DU for total column.
Method 3: SW spectroscopy (limb) – Principle: Scattered radiation in UV and VIS observed in limb
mode with high spectral resolution. Also, missing lines from the spectrum of the Sun, moon or
stars during occultation. Applicable only in LEO.
Method 4: IR spectroscopy (limb) – Principle: Emitted radiation in TIR (~9.7 µm) observed with
high spectral resolution by limb-sounding spectrometers. Applicable only in LEO.
Method 6: DIAL – Principle: Backscattered radiation in a UV, VIS or TIR ozone-absorption band
and a side window by differential absorption lidar. Applicable only in LEO.
Definition: 3D field of dry-air mole fraction of BrO. Required from surface to TOA (layers: LT, HT,
LS, HS&M) – Physical unit: nmol/mol – Uncertainty unit: nmol/mol.
Definition: 3D field of dry-air mole fraction of C 2H2. Required in the troposphere (layers: LT, HT) –
Physical unit: nmol/mol – Uncertainty unit: nmol/mol.
Method 1: TIR spectroscopy (cross-nadir) – Principle: Emitted radiation in TIR (~7.5 and 13.7 µm)
observed with high spectral resolution by cross-nadir spectrometers. Applicable in both LEO
and GEO.
Definition: 3D field of dry-air mole fraction of C 2H6. Required in the troposphere (layers: LT, HT) –
Physical unit: nmol/mol – Uncertainty unit: nmol/mol.
Method 1: TIR spectroscopy (cross-nadir) – Principle: Emitted radiation in TIR (~3.3 and 12 µm)
observed with high spectral resolution by cross-nadir spectrometers. Applicable in both LEO
and GEO.
Definition: 3D field of dry-air mole fraction of CFC–11. Required from surface to TOA (layers: LT,
HT, LS, HS&M) – Physical unit: nmol/mol – Uncertainty unit: nmol/mol.
148 GUIDE TO INSTRUMENTS AND METHODS OF OBSERVATION - VOLUME IV
Method 1: TIR spectroscopy (cross-nadir) – Principle: Emitted radiation in TIR (~9.2 and 11.7 µm)
observed with medium-high spectral resolution by cross-nadir spectrometers. Applicable in both
LEO and GEO.
Method 2: TIR spectroscopy (limb) – Principle: Emitted radiation in TIR (~9.2 and 11.7 µm)
observed with medium-high spectral resolution by limb-sounding spectrometers. Applicable
only in LEO.
Definition: 3D field of dry-air mole fraction of CFC–12. Required from surface to TOA (layers: LT,
HT, LS, HS&M) – Physical unit: nmol/mol – Uncertainty unit: nmol/mol.
Method 1: TIR spectroscopy (cross-nadir) – Principle: Emitted radiation in TIR (~8.8 and 10.8 µm)
observed with medium-high spectral resolution by cross-nadir spectrometers. Applicable in both
LEO and GEO.
Method 2: TIR spectroscopy (limb) – Principle: Emitted radiation in TIR (~8.8 and 10.8 µm)
observed with medium-high spectral resolution by limb-sounding spectrometers. Applicable
only in LEO.
Definition: 3D field of dry-air mole fraction of CH2O. Required in the troposphere (layers: LT, HT)
+ total column – Physical unit: nmol/mol for layers, units of 1.3 · 1015 molecules/cm2 for total
column – Uncertainty unit: nmol/mol for layers, 1.3 · 1015 cm–2 for total column.
Definition: 3D field of dry-air mole fraction of CH4. Required from surface to TOA (layers: LT, HT,
LS, HS&M) + total column – Physical unit: nmol/mol for layers, units of 1.3 · 1015 molecules/
cm2 for total column – Uncertainty unit: nmol/mol for layers, 1.3 · 1015 molecules/cm2 for total
column.
Method 2: TIR spectroscopy (cross-nadir) – Principle: Emitted radiation in TIR (~3.4, 4.3 and
7.7 µm) observed with medium-high spectral resolution by cross-nadir spectrometers.
Applicable in both LEO and GEO.
Method 3: SWIR spectroscopy (limb) – Principle: Scattered radiation in SWIR (~2.3 µm) observed
in limb mode with high spectral resolution. Also, missing lines from the spectrum of the Sun,
moon or stars during occultation. Applicable only in LEO.
Method 4: TIR spectroscopy (limb) – Principle: Emitted radiation in TIR (~3.4, 4.3 and 7.7 µm)
observed with high spectral resolution by limb-sounding spectrometers. Applicable only in LEO.
CHAPTER 5. SPACE-BASED OBSERVATION OF GEOPHYSICAL VARIABLES 149
Definition: 3D field of dry-air mole fraction of ClO. Required from surface to TOA (layers: LT, HT,
LS, HS&M) – Physical unit: nmol/mo – Uncertainty unit: nmol/mol.
Definition: 3D field of dry-air mole fraction of ClONO2. Required from mid-troposphere to TOA
(layers: HT, LS, HS&M) – Physical unit: nmol/mol – Uncertainty unit: nmol/mol.
Method 1: TIR spectroscopy (cross-nadir) – Principle: Emitted radiation in TIR (~5.7, 7.7 and
12.5 µm) observed with high spectral resolution by cross-nadir spectrometers. Applicable in both
LEO and GEO.
Method 2: TIR spectroscopy (limb) – Principle: Emitted radiation in TIR (~5.7, 7.7 and 12.5 µm)
observed with high spectral resolution by limb-sounding spectrometers. Applicable only in LEO.
Definition: 3D field of dry-air mole fraction of CO. Required from surface to low
stratosphere (layers: LT, HT, LS) + total column – Physical unit: nmol/mol for layers, units
of 1.3 · 1015 molecules/cm2 for total column – Uncertainty unit: nmol/mol for layers,
1.3 · 1015 molecules/cm2 for total column.
Method 2: TIR spectroscopy (cross-nadir) – Principle: Emitted radiation in TIR (~4.6 µm) observed
with medium-high spectral resolution by cross-nadir spectrometers. Applicable in both LEO
and GEO.
Method 3: SWIR spectroscopy (limb) – Principle: Scattered radiation in SWIR (~2.3 µm) observed
in limb mode with high spectral resolution. Also, missing lines from the spectrum of the Sun,
moon or stars during occultation. Applicable only in LEO.
Method 4: TIR spectroscopy (limb) – Principle: Emitted radiation in TIR (~4.6 µm) observed with
high spectral resolution by limb-sounding spectrometers. Applicable only in LEO.
Definition: 3D field of dry-air mole fraction of CO2. Required from surface to TOA (layers: LT, HT,
LS, HS&M) + total column – Physical unit: nmol/mol for layers, units of
1.3 · 1015 molecules/cm2 for total column – Uncertainty unit: nmol/mol for layers,
1.3 · 1015 molecules/cm2 for total column.
150 GUIDE TO INSTRUMENTS AND METHODS OF OBSERVATION - VOLUME IV
Method 1: SWIR spectroscopy (cross-nadir) – Principle: Scattered radiation in SWIR (~1.6 and
2.1 µm) observed with high spectral resolution by cross-nadir spectrometers. Applicable in both
LEO and GEO.
Method 2: TIR spectroscopy (cross-nadir) – Principle: Emitted radiation in TIR (~4.3 and 15 µm)
observed with medium-high spectral resolution by cross-nadir spectrometers. Applicable in both
LEO and GEO.
Method 3: SWIR spectroscopy (limb) – Principle: Scattered radiation in SWIR (~1.6 and 2.1 µm)
observed in limb mode with high spectral resolution. Also, missing lines from the spectrum of
the Sun, moon or stars during occultation. Applicable only in LEO.
Method 4: TIR spectroscopy (limb) – Principle: Emitted radiation in TIR (~4.3 and 15 µm) observed
with high spectral resolution by limb-sounding spectrometers. Applicable only in LEO.
Method 5: DIAL – Principle: Backscattered radiation in a CO2 absorption band and a side window
by differential absorption lidar. Several bands are available, around 1.6 and 2.0 µm, for example.
Only total column feasible. Integration over large area and long time necessary to achieve the
required uncertainty (~0.3%). Applicable only in LEO.
Definition: 3D field of dry-air mole fraction of COS. Required from surface to low stratosphere
(layers: LT, HT, LS) – Physical unit: nmol/mol – Uncertainty unit: nmol/mol.
Method 1: TIR spectroscopy (cross-nadir) – Principle: Emitted radiation in TIR (~4.8 and 11.6 µm)
observed with high spectral resolution by cross-nadir spectrometers. Applicable in both LEO
and GEO.
Method 2: TIR spectroscopy (limb) – Principle: Emitted radiation in TIR (~4.8 and 11.6 µm)
observed with high spectral resolution by limb-sounding spectrometers. Applicable only in LEO.
Definition: 3D field of dry-air mole fraction of H2O (intended as a chemical species relevant for
atmospheric chemistry). Required from surface to TOA (layers: LT, HT, LS, HS&M) – Physical
unit: nmol/mol – Uncertainty unit: nmol/mol.
Method 1: SW spectroscopy (cross-nadir) – Principle: Scattered radiation in VIS, NIR and SWIR
observed with high spectral resolution in several bands by cross-nadir spectrometers. Applicable
in both LEO and GEO.
Method 4: GNSS radio-occultation – Principle: Atmospheric refraction of L-band signals from the
GNSS received by a LEO satellite during the occultation phase. Applicable only in LEO.
Method 5: SW spectroscopy (limb) – Principle: Scattered radiation in VIS, NIR and SWIR observed
in limb mode with high spectral resolution. Also, missing lines from the spectrum of the Sun,
moon or stars during occultation. Applicable only in LEO.
CHAPTER 5. SPACE-BASED OBSERVATION OF GEOPHYSICAL VARIABLES 151
Method 6: IR spectroscopy (limb) – Principle: Emitted radiation in TIR (~6.3 µm) observed with
high spectral resolution by limb-sounding spectrometers. Applicable only in LEO.
Method 7: MW/sub-mm spectroscopy (limb) – Principle: Emitted radiation in several bands of the
MW/sub-mm range observed with high spectral resolution by limb-sounding spectrometers.
Applicable only in LEO.
Method 8: DIAL – Principle: Backscattered radiation in a UV, VIS or TIR absorption band and a side
window by differential absorption lidar. Applicable only in LEO.
Definition: 3D field of dry-air mole fraction of HCl. Required from mid-troposphere to TOA (layers:
HT, LS, HS&M) – Physical unit: nmol/mol – Uncertainty unit: nmol/mol.
5.8.16 HDO
Definition: 3D field of dry-air mole fraction of HDO = water vapour (with one hydrogen nucleus
replaced by its deuterium isotope). Required from low stratosphere to TOA (layers: LS and
HS&M) – Physical unit: nmol/mol – Uncertainty unit: nmol/mol.
Definition: 3D field of dry-air mole fraction of HNO3. Required from surface to TOA (layers: LT,
HT, LS, HS&M) + total column – Physical unit: nmol/mol for layers, units of 1.3 · 1015 molecules/
cm2 for total column – Uncertainty unit: nmol/mol for layers, 1.3 · 1015 molecules/cm2 for total
column.
Method 1: IR spectroscopy (cross-nadir) – Principle: Emitted radiation in TIR (~5.9, 7.6 and
11.3 µm) observed with high spectral resolution by cross-nadir spectrometers. Applicable in both
LEO and GEO.
Method 2: IR spectroscopy (limb) – Principle: Emitted radiation in TIR (~5.9, 7.6 and 11.3 µm)
observed with high spectral resolution by limb-sounding spectrometers. Applicable only in LEO.
Definition: 3D field of dry-air mole fraction of N2O. Required from surface to TOA (layers: LT, HT,
LS, HS&M) – Physical unit: nmol/mol – Uncertainty unit: nmol/mol.
Method 2: TIR spectroscopy (cross-nadir) – Principle: Emitted radiation in TIR (~4.5 and 7.7 µm)
observed with medium-high spectral resolution by cross-nadir spectrometers. Applicable in both
LEO and GEO.
Method 3: SWIR spectroscopy (limb) – Principle: Scattered radiation in SWIR (~2.3 µm) observed
in limb mode with high spectral resolution. Also, missing lines from the spectrum of the Sun,
moon or stars during occultation. Applicable only in LEO.
Method 4: TIR spectroscopy (limb) – Principle: Emitted radiation in TIR (~4.5 and 7.7 µm)
observed with high spectral resolution by limb-sounding spectrometers. Applicable only in LEO.
Definition: 3D field of dry-air mole fraction of N2O5. Required in the troposphere (layers: LT, HT) –
Physical unit: nmol/mol – Uncertainty unit: nmol/mol.
Method 1: TIR spectroscopy (cross-nadir) – Principle: Emitted radiation in TIR (~5.8 and 8.0 µm)
observed with high spectral resolution by cross-nadir spectrometers. Applicable in both LEO
and GEO.
Definition: 3D field of dry-air mole fraction of NO. Required from surface to TOA (layers: LT, HT,
LS, HS&M) – Physical unit: nmol/mol – Uncertainty unit: nmol/mol.
Method 4: IR spectroscopy (limb) – Principle: Emitted radiation in TIR (~5.3 µm) observed with
high spectral resolution by limb-sounding spectrometers. Applicable only in LEO.
Definition: 3D field of dry-air mole fraction of NO2. Required from surface to TOA (layers: LT, HT,
LS, HS&M) + total column – Physical unit: nmol/mol for layers, units of 1.3 · 1015 molecules/cm2
for total column – Uncertainty unit: nmol/mol for layers, 1.3 · 1015 cm–2 for total column.
Method 3: SW spectroscopy (limb) – Principle: Scattered radiation in UV and VIS observed in limb
mode with high spectral resolution. Also, missing lines from the spectrum of the Sun, moon or
stars during occultation. Applicable only in LEO.
Method 4: IR spectroscopy (limb) – Principle: Emitted radiation in TIR (~6.1 µm) observed with
high spectral resolution by limb-sounding spectrometers. Applicable only in LEO.
Definition: 3D field of dry-air mole fraction of OH. Required from surface to TOA (layers: LT, HT,
LS, HS&M) – Physical unit: nmol/mol – Uncertainty unit: nmol/mol.
Method 2: FIR spectroscopy (limb) – Principle: Emitted radiation in the Far IR (several lines in the
range 28–182 µm, best at ~84 µm) observed with high spectral resolution by limb-sounding
spectrometers. Applicable only in LEO.
Definition: 3D field of dry-air mole fraction of peroxyacetyl nitrate. Required in the troposphere
(layers: LT, HT) – Physical unit: nmol/mol – Uncertainty unit: nmol/mol.
Method 1: TIR spectroscopy (cross-nadir) – Principle: Emitted radiation in TIR (~5.7, 8.6 and
12.5 µm) observed with high spectral resolution by cross-nadir spectrometers. Applicable only
in LEO.
Definition: 3D field of polar stratospheric cloud occurrence. Required in the lower stratosphere
(layer: LS) – Uncertainty expressed as hit rate (HR) and false-alarm rate (FAR).
Method 2: SW spectroscopy (limb) – Principle: Scattered radiation in UV and VIS observed in limb
mode with moderate spectral resolution. Applicable only in LEO.
Method 3: IR spectroscopy (limb) – Principle: Emitted radiation in TIR observed with moderate
spectral resolution by limb-sounding spectrometers. Applicable only in LEO.
Method 4: Backscatter lidar – Principle: Backscattered radiation in UV, VIS or NIR by lidar.
Applicable only in LEO.
Definition: 3D field of dry-air mole fraction of SF6. Required from low stratosphere to TOA (layers:
LS and HS&M) – Physical unit: nmol/mol – Uncertainty unit: nmol/mol.
Method 1: TIR spectroscopy (cross-nadir) – Principle: Emitted radiation in TIR (~10.5 µm) observed
with high spectral resolution by cross-nadir spectrometers. Applicable in both LEO and GEO.
Method 2: TIR spectroscopy (limb) – Principle: Emitted radiation in TIR (~10.5 µm) observed with
high spectral resolution by limb-sounding spectrometers. Applicable only in LEO.
154 GUIDE TO INSTRUMENTS AND METHODS OF OBSERVATION - VOLUME IV
Definition: 3D field of dry-air mole fraction of SO2. Required from surface to lower
stratosphere (layers: LT, HT, LS) + total column – Physical unit: nmol/mol for layers, units of
1.3 · 1015 molecules/cm2 for total column – Uncertainty unit: nmol/mol for layers, 1.3 · 1015 cm–2
for total column.
Method 2: TIR spectroscopy (cross-nadir) – Principle: Emitted radiation in TIR (~7.3 and 8.6 µm)
observed with medium-high spectral resolution by cross-nadir spectrometers. Applicable in both
LEO and GEO.
This theme comprises variables that characterize space weather. The variables relevant to this
theme are classified below according to three categories:
(b) Sun–Earth interplanetary space, dominated by the solar wind (Table 5.9);
Solar radio flux Radio flux integrated over the solar disk W · m–2 · Hz–1
spectrum
Solar radio flux image Radio flux received from the solar disk W · m–2 · Hz–1 · arcsec–2
Solar electric field Map of the electric field in the photosphere mV · m–1 · arcsec–2
Solar corona image Image of the corona surrounding the Sun W · m–2 · arcsec–2
CHAPTER 5. SPACE-BASED OBSERVATION OF GEOPHYSICAL VARIABLES 155
Heavy ions [2(He) < Z ≤ 26(Fe)] Angular flux energy and mass particles · m–2 · s–1 · sr–1
spectrum (MeV/nucleon)–1
Integral directional flux particles · m–2 · s–1 · sr–1
Ionospheric total electron Number of electrons between two points Total electron
content content units
(TECU)
Electric field Magnitude and direction of the Earth’s electric field mV · m
Radiation dose rate 3D field of the dose rate of energetic particles mSv · h–1
In the following sections, some details are given for a few variables that are relevant to the
ionosphere and magnetosphere.
Definition: Number of electrons along a path between two points. Observed under different
viewing angles so as to generate vertical profiles by tomography. Required in the ionosphere and
plasmasphere – Physical unit: electrons/m2; practical unit: TECU = 1016 electrons/m2 – Uncertainty
unit: %.
156 GUIDE TO INSTRUMENTS AND METHODS OF OBSERVATION - VOLUME IV
Method 2: Radar altimetry – Principle: Differential phase delay between signals from dual-
frequency radar altimeter (~13 GHz and ~3 or 5 GHz). Phase rotation measurement, primarily
needed to correct the altimeter ranging measurement, is also used to infer the column-integrated
total electron content. Applicable only in LEO.
Method 3: GPS–LEO signal phase delay – Principle: Differential phase delay between signals
from two-frequency GPS transmitters (~1.2 and 1.6 GHz) and a receiver in LEO using GPS for
navigation. In principle, any satellite equipped with a GPS navigation system is suitable. The
information refers to the topside ionosphere and plasmasphere, namely the layer between the
satellite altitude and the GPS altitude (~20 000 km). Applicable only in LEO.
Definition: 3D distribution of the electron density. Required in the ionosphere and plasmasphere –
Physical unit: electrons/m3 – Uncertainty unit: %.
Definition: Magnitude and direction of the Earth’s magnetic field. Indicative of the degree of
geomagnetic disturbance within the magnetosphere, and also in the Earth’s interior. Required in
the magnetosphere – Physical unit: nT (1 tesla = 10 4 gauss) – Uncertainty: nT.
Method 1: Magnetometry – Principle: more magnetometers for in situ measurement along the
orbit as the satellite moves. Applicable in LEO, in GEO and in highly elliptical orbits.
Definition: Magnitude and direction of the Earth’s electric field. Required in the ionosphere –
Physical unit: mV · m–1 – Uncertainty: mV · m–1.
Method 1: Ion drift – Principle: Measurement of magnitude and direction of the incoming ion
flux. The electric field is derived from the relationship between electric field, measured ion drift
velocity and measured magnetic field strength. In situ measurement along the orbit as the
satellite moves. Applicable in LEO and in highly elliptical orbits.
ANNEX. ACHIEVABLE QUALITY OF SATELLITE PRODUCTS
This annex indicates the achievable quality in terms of RMS error,1 horizontal resolution (Δx),
vertical resolution (Δz) and observing cycle (Δt), with assumptions made on the number of
satellites needed for the quoted observing cycle Δt, and the main possible observing conditions
or limitations. These quality estimates are based on the characteristics of state-of-the art
instruments that are being developed at the time of writing this Guide, and expected to be
operational by 2020. This estimate is made for each applicable remote-sensing principle for
geophysical variables of the eight following themes:
1
Note that RMS error used in the present volume corresponds approximately to an expanded uncertainty with a
coverage factor of k = 1, while in the rest of this Guide a coverage factor of k = 2 is generally used (see Volume 1,
Chapter 1, 1.6.3.3 and the Evaluation of Measurement Data – Guide to the Expression of Uncertainty in Measurement
(JCGM 100:2008).
158 GUIDE TO INSTRUMENTS AND METHODS OF OBSERVATION - VOLUME IV
Table 5.A.1.1. Estimated potential quality of product “Atmospheric temperature” (by 2020)
Uncertainty Δx Δz Δt Number
Layer Orbit Technique Conditions
(RMS) (km) (km) (h) of sats
Troposphere LEO IR spectroscopy 1K 20 1 4 3 Clear-air
(at ~500 hPa)
LEO MW/sub-mm 4K 30 4 4 3 –
radiometry
Table 5.A.1.2. Estimated potential quality of product “Specific humidity” (by 2020)
Uncertainty Δx Δz Δt Number
Layer Orbit Technique Conditions
(RMS) (km) (km) (h) of sats
Troposphere LEO IR spectroscopy 10% 20 1.5 4 3 Clear-air
(at ~500 hPa)
GEO IR spectroscopy 10% 20 2 0.5 6 Clear-air
Table 5.A.1.3. Estimated potential quality of product “Wind (horizontal)” (by 2020)
Uncertainty Δx Δz Δt Number
Layer Orbit Technique Conditions
(RMS) (km) (km) (h) of sats
Troposphere LEO Doppler lidar 1 m · s–1 50 0.5 180 1 Clear-air
(at ~500 hPa) (non-scanning)
Uncertainty Δx Δz Δt Number
Layer Orbit Technique Conditions
(RMS) (km) (km) (h) of sats
Radar
LEO 2 m · s–1 20 – 12 3 Over sea, all weather
scatterometer
Polarimetric MW
LEO 3 m · s–1 10 – 8 3 Over sea, all weather
radiometry
Surface
Over sea, all weather,
LEO MW imagery 3 m · s–1 10 – 8 3
speed only
Uncertainty Δx Δz Δt Number
Layer Orbit Technique Conditions
(RMS) (km) (km) (h) of sats
Backscatter lidar
LEO 0.1 km 50 – 360 1 Clear-air
(non-scanning)
From GNSS
LEO 0.3 km 300 – 12 12 All weather
sounding
CHAPTER 5. SPACE-BASED OBSERVATION OF GEOPHYSICAL VARIABLES 161
Uncertainty Δx Δz Δt Number
Layer Orbit Technique Conditions
(RMS) (km) (km) (h) of sats
Backscatter lidar
LEO 0.1 km 50 – 360 1 Clear-air
(non-scanning)
Uncertainty Δx Δz Δt Number
Layer Orbit Technique Conditions
(RMS) (km) (km) (h) of sats
LEO From IR sounding 2K 20 – 4 3 Clear-air
Cloud-top height Cloud optical depth Cloud-ice effective radius Precipitation intensity at
surface (liquid or solid)
Table 5.A.2.1. Estimated potential quality of product “Cloud-top temperature” (by 2020)
Uncertainty Δx Δz Δt Number
Layer Orbit Technique Conditions
(RMS) (km) (km) (h) of sats
LEO IR radiometry 2K 1 – 4 3 –
Table 5.A.2.2. Estimated potential quality of product “Cloud-top height” (by 2020)
Uncertainty Δx Δz Δt Number
Layer Orbit Technique Conditions
(RMS) (km) (km) (h) of sats
LEO IR radiometry 0.5 km 1 – 4 3 –
Backscatter lidar
LEO 0.1 km 50 – 360 1 Multi-orbit
N/A (non-scanning)
Cloud radar
LEO 0.3 km 50 – 360 1 Multi-orbit
(non-scanning)
A-band
LEO 0.3 km 4 – 8 3 Daylight
spectroscopy
A-band
GEO 0.3 km 8 – 0.25 6 Daylight
spectroscopy
Table 5.A.2.3. Estimated potential quality of product “Cloud type” (by 2020)
Uncertainty Δx Δz Δt Number
Layer Orbit Technique Conditions
(RMS) (km) (km) (h) of sats
Degraded at night
LEO VIS/IR radiometry 10 classes 4 – 4 3
(no VIS)
N/A
Degraded at night
GEO VIS/IR radiometry 8 classes 12 – 0.1 1
(no VIS)
Table 5.A.2.4. Estimated potential quality of product “Cloud cover” (by 2020)
Uncertainty Δx Δz Δt Number
Layer Orbit Technique Conditions
(RMS) (km) (km) (h) of sats
Troposphere LEO From IR sounding 10% 10 6 4 3 Within sounded
IFOV
Table 5.A.2.5. Estimated potential quality of product “Cloud-base height” (by 2020)
Uncertainty Δx Δz Δt Number
Layer Orbit Technique Conditions
(RMS) (km) (km) (h) of sats
Cloud radar
N/A LEO 0.3 km 50 – 360 1 Multi-orbit
(non-scanning)
Table 5.A.2.6. Estimated potential quality of product “Cloud optical depth” (by 2020)
Uncertainty Δx Δz Δt Number
Layer Orbit Technique Conditions
(RMS) (km) (km) (h) of sats
Backscatter lidar
LEO 0.1 50 – 360 1 Multi-orbit
(non-scanning)
Table 5.A.2.7. Estimated potential quality of product “Cloud liquid water” (by 2020)
Uncertainty Δx Δz Δt Number
Layer Orbit Technique Conditions
(RMS) (km) (km) (h) of sats
Troposphere LEO Cloud radar 10% 50 0.3 360 1 Multi-orbit
(non-scanning)
Uncertainty Δx Δz Δt Number
Layer Orbit Technique Conditions
(RMS) (km) (km) (h) of sats
Troposphere LEO Cloud radar 3 µm 50 0.3 360 1 Non-precipitating
(non-scanning) clouds
LEO MW/sub-mm 10 µm 20 – 3 8 –
sounding (GPM)
Table 5.A.2.9. Estimated potential quality of product “Cloud ice” (by 2020)
Uncertainty Δx Δz Δt Number
Layer Orbit Technique Conditions
(RMS) (km) (km) (h) of sats
Troposphere LEO Cloud radar 10% 50 0.3 360 1 Non-precipitating
(non-scanning) clouds
Uncertainty Δx Δz Δt Number
Layer Orbit Technique Conditions
(RMS) (km) (km) (h) of sats
Troposphere LEO Cloud radar 3 µm 50 0.3 360 1 Non-precipitating
(non-scanning) clouds
LEO MW/sub-mm 10 µm 20 – 3 8 –
sounding (GPM)
Uncertainty Δx Δz Δt Number
Layer Orbit Technique Conditions
(RMS) (km) (km) (h) of sats
LEO Precipitation radar 0.3 km 5 – 120 1 –
Uncertainty Δx Δz Δt Number
Layer Orbit Technique Conditions
(RMS) (km) (km) (h) of sats
LEO Precipitation radar 0.3 km 5 – 120 1 –
Uncertainty Δx Δz Δt Number
Layer Orbit Technique Conditions
(RMS) (km) (km) (h) of sats
LEO Precipitation radar 10% 5 0.3 120 1 –
MW/sub-mm Heavily
LEO 30% 20 3 3 8 (GPM)
Troposphere sounding model-aided
MW/sub-mm Heavily
GEO 30% 30 3 0.5 6
sounding model-aided
Uncertainty Δx Δz Δt Number
Layer Orbit Technique Conditions
(RMS) (km) (km) (h) of sats
LEO Precipitation radar 0.5 mm h–1 5 – 120 1 –
Heavily
LEO MW/sub-mm sounding 1 mm h–1 10 – 3 8 (GPM)
model-aided
Heavily
GEO MW/sub-mm sounding 1.5 mm h–1 10 – 0.5 6
Surface model-aided
Convection
GEO VIS/IR radiometry 5 mm h–1 10 – 0.1 6
only
Product from
GEO LEO/MW + GEO/IR fusion 2.5 mm h–1 10 – 0.1 6
data-fusion
Uncertainty Δx Δz Δt Number
Layer Orbit Technique Conditions
(RMS) (km) (km) (h) of sats
LEO/MW + Product from
GEO 5 mm 10 – 3 3 (LEO) + 6 (GEO)
GEO/IR fusion data-fusion
Surface
MW/sub-mm Heavily
GEO 2 mm 10 – 3 6
sounding model-aided
Table 5.A.2.16. Estimated potential quality of product “Lightning detection” (by 2020)
Uncertainty Δx Δz Δt Number
Layer Orbit Technique Conditions
(RMS) (km) (km) (h) of sats
LEO Lightning mapping 0.10/0.95 FAR/HR 5 – 12 3 –
N/A
GEO Lightning mapping 0.15/0.90 FAR/HR 10 – 0.01 6 –
168 GUIDE TO INSTRUMENTS AND METHODS OF OBSERVATION - VOLUME IV
Table 5.A.3. Geophysical variables considered under the theme “Aerosol and radiation”
Aerosol optical depth Upward spectral radiance at TOA Earth’s surface albedo
Aerosol effective radius Upward SW irradiance at TOA Upward LW irradiance at Earth’s surface
Table 5.A.3.1. Estimated potential quality of product “Aerosol optical depth” (by 2020)
Uncertainty Δx Δz Δt Number
Layer Orbit Technique Conditions
(RMS) (km) (km) (h) of sats
LEO Backscatter lidar 0.01 50 – 360 1 Clear-air
(non-scanning)
Total
column
GEO SW spectroscopy 0.04 20 – 1 6 Clear-air, daylight
Table 5.A.3.2. Estimated potential quality of product “Aerosol concentration” (by 2020)
Uncertainty Δx Δz Δt Number
Layer Orbit Technique Conditions
(RMS) (km) (km) (h) of sats
Troposphere LEO Backscatter lidar 1% 50 0.1 360 1 Clear-air
(at ~500 hPa) (non-scanning)
Table 5.A.3.3. Estimated potential quality of product “Aerosol effective radius” (by 2020)
Uncertainty Δx Δz Δt Number
Layer Orbit Technique Conditions
(RMS) (km) (km) (h) of sats
Troposphere LEO Backscatter lidar 0.2 µm 50 0.3 360 1 Clear-air
(non-scanning)
Table 5.A.3.4. Estimated potential quality of product “Aerosol type” (by 2020)
Uncertainty Δx Δz Δt Number
Layer Orbit Technique Conditions
(RMS) (km) (km) (h) of sats
Troposphere LEO Backscatter lidar 6 classes 50 0.3 360 1 Clear-air
(non-scanning)
Table 5.A.3.5. Estimated potential quality of product “Volcanic ash” (by 2020)
Uncertainty Δx Δz Δt Number
Layer Orbit Technique Conditions
(RMS) (km) (km) (h) of sats
Troposphere LEO Backscatter lidar 1% 50 0.1 360 1 Clear-air
(non-scanning)
Uncertainty Δx Δz Δt Number
Layer Orbit Technique Conditions
(RMS) (km) (km) (h) of sats
Multi-orbit
TOA LEO Cavity radiometer 0.2 W · m–2 N/A – 24 1
integration
Uncertainty Δx Δz Δt Number
Layer Orbit Technique Conditions
(unit) (km) (km) (h) of sats
50 – 720 1 Non-scanning
TOA LEO SW+LW spectroscopy 100 SNR
10 – 168 1 Limited swath
172 GUIDE TO INSTRUMENTS AND METHODS OF OBSERVATION - VOLUME IV
Uncertainty Δx Δz Δt Number
Layer Orbit Technique Conditions
(RMS) (km) (km) (h) of sats
LEO Broadband radiometry 4 W · m–2 20 – 4 3 –
TOA
GEO Broadband radiometry 4 W · m–2 30 – 0.25 6 –
Uncertainty Δx Δz Δt Number
Layer Orbit Technique Conditions
(RMS) (km) (km) (h) of sats
LEO Broadband radiometry 10 W · m–2 20 – 4 3 Model-aided
TOA
GEO Broadband radiometry 15 W · m–2 30 – 0.25 6 Model-aided
Uncertainty Δx Δz Δt Number
Layer Orbit Technique Conditions
(RMS) (km) (km) (h) of sats
LEO SW radiometry 5% 4 – 4 3 –
TOA
GEO SW radiometry 7% 8 – 0.1 6 –
Uncertainty Δx Δz Δt Number
Layer Orbit Technique Conditions
(RMS) (km) (km) (h) of sats
LEO From IR/MW sounding 10 W · m–2 20 – 4 3 Model-aided
Surface
GEO From IR/MW sounding 10 W · m–2 20 – 0.5 6 Model-aided
Uncertainty Δx Δz Δt Number
Layer Orbit Technique Conditions
(RMS) (km) (km) (h) of sats
Clear-air,
LEO SW radiometry 20 W · m–2 4 – 4 3
model-aided
Surface
Clear-air,
GEO SW radiometry 30 W · m–2 8 – 0.1 6
model-aided
CHAPTER 5. SPACE-BASED OBSERVATION OF GEOPHYSICAL VARIABLES 173
Table 5.A.3.13. Estimated potential quality of product “Earth’s surface albedo” (by 2020)
Uncertainty Δx Δz Δt Number
Layer Orbit Technique Conditions
(RMS) (km) (km) (h) of sats
Multi-view SW Clear-air,
LEO 1% 10 – 168 1
radiometry model-aided
Clear-air, heavily
Surface LEO VIS radiometry 3% 4 – 168 3
model-aided
Clear-air, heavily
GEO VIS radiometry 5% 8 – 72 6
model-aided
Uncertainty Δx Δz Δt Number
Layer Orbit Technique Conditions
(RMS) (km) (km) (h) of sats
Surface LEO Multi-view SW radiometry 3% 8 – 168 1 Clear-air
Uncertainty Δx Δz Δt Number
Layer Orbit Technique Conditions
(RMS) (km) (km) (h) of sats
Clear-air,
LEO Broadband radiometry 15 W · m–2 20 – 4 3
model-aided
Surface
Clear-air,
GEO Broadband radiometry 15 W · m–2 30 – 0.25 6
model-aided
Uncertainty Δx Δz Δt Number
Layer Orbit Technique Conditions
(RMS) (km) (km) (h) of sats
LEO IR radiometry 3% 4 – 168 3 Clear-air
Uncertainty Δx Δz Δt Number
Layer Orbit Technique Conditions
(RMS) (km) (km) (h) of sats
Clear-air,
LEO VIS radiometry 10 W · m–2 4 – 4 3
model-aided
Surface
Clear-air,
GEO VIS radiometry 10 W · m–2 8 – 0.1 6
model-aided
174 GUIDE TO INSTRUMENTS AND METHODS OF OBSERVATION - VOLUME IV
Uncertainty Δx Δz Δt Number
Layer Orbit Technique Conditions
(RMS) (km) (km) (h) of sats
Clear-air,
LEO VIS radiometry 10% 4 – 4 3
model-aided
Surface
Clear-air,
GEO VIS radiometry 10% 8 – 0.1 6
model-aided
Table 5.A.4. Geophysical variables considered under the theme “Ocean and sea ice”
Ocean chlorophyll Oil-spill cover Coastal sea level (tide) Wave directional-energy
concentration frequency spectrum
Colour dissolved organic Sea-surface temperature Significant wave height Sea-ice cover
matter
Ocean diffuse attenuation Ocean dynamic Dominant wave period Sea-ice type
coefficient topography
Uncertainty Δx Δz Δt Number
Layer Orbit Technique Conditions
(RMS) (km) (km) (h) of sats
Clear-air, daylight,
LEO VIS radiometry 0.1 mg · m–3 4 – 8 3
model-aided
Surface
Clear-air, daylight,
GEO VIS radiometry 0.2 mg · m–3 8 – 0.25 6
model-aided
Uncertainty Δx Δz Δt Number
Layer Orbit Technique Conditions
(RMS) (km) (km) (h) of sats
Clear-air, daylight,
LEO VIS radiometry 0.01 m–1 4 – 8 3
model-aided
Surface
Clear-air, daylight,
GEO VIS radiometry 0.02 m–1 8 – 0.25 6
model-aided
CHAPTER 5. SPACE-BASED OBSERVATION OF GEOPHYSICAL VARIABLES 175
Uncertainty Δx Δz Δt Number
Layer Orbit Technique Conditions
(RMS) (km) (km) (h) of sats
Clear-air, daylight,
LEO VIS radiometry 0.05 g · m–3 4 – 8 3
model-aided
Surface
Clear-air, daylight,
GEO VIS radiometry 0.1 g · m–3 8 – 0.25 6
model-aided
Uncertainty Δx Δz Δt Number
Layer Orbit Technique Conditions
(RMS) (km) (km) (h) of sats
Clear-air, daylight,
LEO VIS radiometry 0.5 m–1 4 – 8 3
model-aided
Surface
Clear-air, daylight,
GEO VIS radiometry 1 m–1 8 – 0.25 6
model-aided
Table 5.A.4.5. Estimated potential quality of product “Oil-spill cover” (by 2020)
Uncertainty Δx Δz Δt Number
Layer Orbit Technique Conditions
(RMS) (km) (km) (h) of sats
LEO VIS/NIR radiometry 15% 4 – 8 3 Clear-air, daylight
Table 5.A.4.6. Estimated potential quality of product “Sea-surface temperature” (by 2020)
Uncertainty Δx Δz Δt Number
Layer Orbit Technique Conditions
(RMS) (km) (km) (h) of sats
LEO IR radiometry 0.4 K 8 – 4 3 Clear-air
Table 5.A.4.7. Estimated potential quality of product “Sea-surface salinity” (by 2020)
Uncertainty Δx Δz Δt Number
Layer Orbit Technique Conditions
(RMS) (km) (km) (h) of sats
L-band MW All weather, space-
Surface LEO 0.3 PSU 200 – 240 1
radiometry time integrated
176 GUIDE TO INSTRUMENTS AND METHODS OF OBSERVATION - VOLUME IV
Uncertainty Δx Δz Δt Number
Layer Orbit Technique Conditions
(RMS) (km) (km) (h) of sats
Radar altimetry
Surface LEO 3 cm 50 – 240 2 All weather
(non-scanning)
Table 5.A.4.9. Estimated potential quality of product “Coastal sea level (tide)” (by 2020)
Uncertainty Δx Δz Δt Number
Layer Orbit Technique Conditions
(RMS) (km) (km) (h) of sats
Radar altimetry
Surface LEO 3 cm 50 – 240 2 All weather
(non-scanning)
Uncertainty Δx Δz Δt Number
Layer Orbit Technique Conditions
(RMS) (km) (km) (h) of sats
Radar altimetry
LEO 0.1 m 50 – 240 2 All weather
Surface (non-scanning)
Uncertainty Δx Δz Δt Number
Layer Orbit Technique Conditions
(RMS) (km) (km) (h) of sats
Surface LEO From SAR spectra 10 degrees 50 – 240 2 All weather
Uncertainty Δx Δz Δt Number
Layer Orbit Technique Conditions
(RMS) (km) (km) (h) of sats
Surface LEO From SAR spectra 10 s 50 – 240 2 All weather
Uncertainty Δx Δz Δt Number
Layer Orbit Technique Conditions
(RMS) (km) (km) (h) of sats
Surface LEO From SAR spectra 0.1 m2 · Hz–1 · rad–1 50 – 240 2 All weather
CHAPTER 5. SPACE-BASED OBSERVATION OF GEOPHYSICAL VARIABLES 177
Table 5.A.4.14. Estimated potential quality of product “Sea-ice cover” (by 2020)
Uncertainty Δx Δz Δt Number
Layer Orbit Technique Conditions
(RMS) (km) (km) (h) of sats
LEO VIS/IR radiometry 10% 5 – 48 3 Clear-air
Table 5.A.4.15. Estimated potential quality of product “Sea-ice thickness” (by 2020)
Uncertainty Δx Δz Δt Number
Layer Orbit Technique Conditions
(RMS) (km) (km) (h) of sats
Radar altimetry
LEO 25 cm 30 – 720 1 All weather
(non-scanning)
Table 5.A.4.16. Estimated potential quality of product “Sea-ice type” (by 2020)
Uncertainty Δx Δz Δt Number
Layer Orbit Technique Conditions
(RMS) (km) (km) (h) of sats
LEO Radar scatterometry 5 classes 20 – 12 3 All weather
Land surface temperature Leaf area index Snow status (wet/dry) Land surface topography
Soil moisture Fire fractional cover Snow water equivalent Glacier topography
(in the roots region)
Uncertainty Δx Δz Δt Number
Layer Orbit Technique Conditions
(RMS) (km) (km) (km) of sats
LEO IR radiometry 2K 8 – 4 3 Clear-air
Table 5.A.5.2. Estimated potential quality of product “Soil moisture at surface” (by 2020)
Uncertainty Δx Δz Δt Number
Layer Orbit Technique Conditions
(RMS) (km) (km) (h) of sats
All weather,
LEO MW radiometry 0.05 m3 · m–3 30 – 8 3
vegetation-sensitive
All weather,
LEO Radar scatterometry 0.05 m3 · m–3 20 – 36 1
vegetation-sensitive
All weather,
Surface LEO SAR imagery 0.1 m3 · m–3 0.1 – 360 2
vegetation-sensitive
Clear-air,
LEO VIS/IR radiometry 0.5 m3 · m–3 4 – 4 3
vegetation-sensitive
Clear-air,
GEO VIS/IR radiometry 0.5 m3 · m–3 12 – 0.1 6
vegetation-sensitive
Uncertainty Δx Δz Δt Number
Layer Orbit Technique Conditions
(RMS) (km) (km) (h) of sats
L-band MW All weather,
LEO 0.05 m3 · m–3 50 0.0001 72 1
radiometry model-aided
Surface
L-band SAR All weather,
LEO 0.1 m3 · m–3 0.1 0.0001 1 440 1
imagery model-aided
Uncertainty Δx Δz Δt Number
Layer Orbit Technique Conditions
(RMS) (km) (km) (h) of sats
High-resolution
LEO 10% 1 – 168 4 Clear-air, daylight
Surface optical imagery
Table 5.A.5.5. Estimated potential quality of product “Vegetated type” (by 2020)
Uncertainty Δx Δz Δt Number
Layer Orbit Technique Conditions
(RMS) (km) (km) (h) of sats
High-resolution
LEO 20 classes 0.1 – 2 160 4 Clear-air, daylight
Surface optical imagery
Table 5.A.5.6. Estimated potential quality of product “Leaf area index” (by 2020)
Uncertainty Δx Δz Δt Number
Layer Orbit Technique Conditions
(RMS) (km) (km) (h) of sats
LEO SW radiometry 10% 2 – 168 3 Clear-air, daylight
High-resolution
LEO 10% 0.1 – 168 4 Clear-air, daylight
optical imagery
Uncertainty Δx Δz Δt Number
Layer Orbit Technique Conditions
(RMS) (km) (km) (h) of sats
LEO VIS/NIR radiometry 5% 2 – 168 3 Clear-air, daylight
Table 5.A.5.8. Estimated potential quality of product “Fire fractional cover” (by 2020)
Uncertainty Δx Δz Δt Number
Layer Orbit Technique Conditions
(RMS) (km) (km) (h) of sats
Degraded at night
LEO VIS/IR radiometry 12% 4 – 4 3
(no VIS)
Degraded at night
GEO VIS/IR radiometry 25% 12 – 0.1 6
Surface (no VIS)
High-resolution
LEO 10% 0.1 – 168 4 Clear-air, daylight
optical imagery
Table 5.A.5.9. Estimated potential quality of product “Fire temperature” (by 2020)
Uncertainty Δx Δz Δt Number
Layer Orbit Technique Conditions
(RMS) (km) (km) (h) of sats
LEO IR radiometry 10 K 2 – 4 3 Clear-air
Surface
GEO IR radiometry 20 K 6 – 0.1 6 Clear-air
180 GUIDE TO INSTRUMENTS AND METHODS OF OBSERVATION - VOLUME IV
Table 5.A.5.10. Estimated potential quality of product “Fire radiative power” (by 2020)
Uncertainty Δx Δz Δt Number
Layer Orbit Technique Conditions
(RMS) (km) (km) (h) of sats
LEO IR radiometry 10 kW · m–2 2 – 4 3 Clear-air
Surface
GEO IR radiometry 20 kW · m–2 6 – 0.1 6 Clear-air
Table 5.A.5.11. Estimated potential quality of product “Snow status (wet/dry)” (by 2020)
Uncertainty Δx Δz Δt Number
Layer Orbit Technique Conditions
(RMS) (km) (km) (h) of sats
LEO MW radiometry 0.15/0.90 FAR/HR 10 – 3 8 (GPM) All weather
Surface
LEO SAR imagery 0.25/0.80 FAR/HR 1 – 360 2 All weather
Table 5.A.5.12. Estimated potential quality of product “Snow cover” (by 2020)
Uncertainty Δx Δz Δt Number
Layer Orbit Technique Conditions
(RMS) (km) (km) (h) of sats
LEO VIS/IR radiometry 10% 5 – 48 3 Clear-air
High-resolution
LEO 1% 1 – 168 4 Clear-air, daylight
optical imagery
Table 5.A.5.13. Estimated potential quality of product “Snow water equivalent” (by 2020)
Uncertainty Δx Δz Δt Number
Layer Orbit Technique Conditions
(RMS) (km) (km) (h) of sats
LEO MW radiometry 20 mm 10 – 8 3 All weather
Table 5.A.5.14. Estimated potential quality of product “Soil type” (by 2020)
Uncertainty Δx Δz Δt Number
Layer Orbit Technique Conditions
(RMS) (km) (km) (h) of sats
High-resolution
LEO 20 classes 0.01 – 8 760 4 Clear-air, daylight
Surface optical imagery
Table 5.A.5.15. Estimated potential quality of product “Land cover” (by 2020)
Uncertainty Δx Δz Δt Number
Layer Orbit Technique Conditions
(RMS) (km) (km) (h) of sats
High-resolution
LEO 20 classes 0.01 – 8 760 4 Clear-air, daylight
Surface optical imagery
Uncertainty Δx Δz Δt Number
Layer Orbit Technique Conditions
(RMS) (km) (km) (h) of sats
High-resolution Clear-air,
LEO 2m 0.01 – 8 760 4
VIS stereoscopy daylight
Lidar altimetry
LEO 0.1 m 0.1 – 43 800 1 Clear-air
(non-scanning)
Table 5.A.5.17. Estimated potential quality of product “Glacier cover” (by 2020)
Uncertainty Δx Δz Δt Number
Layer Orbit Technique Conditions
(RMS) (km) (km) (h) of sats
High-resolution Clear-air,
LEO 10% 0.01 – 8 760 4
Surface optical imagery daylight
Table 5.A.5.18. Estimated potential quality of product “Glacier topography” (by 2020)
Uncertainty Δx Δz Δt Number
Layer Orbit Technique Conditions
(RMS) (km) (km) (h) of sats
Surface LEO SAR interferometry 100 cm 0.01 – 8 760 2 All weather
6. SOLID EARTH
Table 5.A.6. Geophysical variables considered under the theme “Solid Earth”
Uncertainty Δx Δz Δt Number
Layer Orbit Technique Conditions
(RMS) (km) (km) (h) of sats
Radar altimetry All weather,
LEO 10 cm 500 – 8 760 2
(non-scanning) model-aided
Surface
All weather, heavily
LEO Gravity field 1 cm 100 – 17 520 1
model-aided
Uncertainty Δx Δz Δt Number
Layer Orbit Technique Conditions
(RMS) (km) (km) (h) of sats
LEO Laser ranging 2 cm 500 – 8 760 5 Night time, clear-air
Surface
LEO GPS receiver 2 cm 100 – 8 760 24 All weather
182 GUIDE TO INSTRUMENTS AND METHODS OF OBSERVATION - VOLUME IV
Uncertainty Δx Δz Δt Number
Layer Orbit Technique Conditions
(RMS) (km) (km) (h) of sats
LEO Laser ranging 2 mm · y–1 500 – 8 760 5 Night time, clear-air
Surface
LEO GPS receiver 2 mm · y–1 100 – 8 760 24 All weather
Table 5.A.6.4. Estimated potential quality of product “Gravity field” (by 2020)
Uncertainty Δx Δz Δt Number
Layer Orbit Technique Conditions
(RMS) (km) (km) (h) of sats
Orbit to change
LEO Gradiometry 2 mGal 300 – 8 760 1
Orbit during mission
height Orbit to change
LEO Sat–sat ranging 2 mGal 300 – 8 760 2
during mission
Table 5.A.6.5. Estimated potential quality of product “Gravity gradients” (by 2020)
Uncertainty Δx Δz Δt Number
Layer Orbit Technique Conditions
(RMS) (km) (km) (h) of sats
Orbit to change
LEO Gradiometry 0.1 E 300 – 8 760 1
Orbit during mission
height Orbit to change
LEO Sat–sat ranging 0.1 E 300 – 8 760 2
during mission
7. ATMOSPHERIC CHEMISTRY
Table 5.A.7. Geophysical variables considered under the theme “Atmospheric chemistry”
Peroxyacetyl
O3 C 2H2 CFC–11 CH2O ClO CO COS HCl HNO3 N2O5 NO2 SF6
nitrate
Polar stratospheric
BrO C 2H6 CFC–12 CH4 ClONO2 CO2 H2O HDO N2O NO OH SO2
cloud occurrence
CHAPTER 5. SPACE-BASED OBSERVATION OF GEOPHYSICAL VARIABLES 183
Uncertainty Δx Δz Δt Number
Layer Orbit Technique Conditions
(RMS) (km) (km) (h) of sats
Troposphere LEO SW spectroscopy 10 nmol mol–1 20 3 8 3 Clear-air,
(at ~500 hPa) (cross-nadir) daylight
Uncertainty Δx Δz Δt Number
Layer Orbit Technique Conditions
(RMS) (km) (km) (h) of sats
Troposphere LEO UV spectroscopy 40 nmol · mol–1 50 5 8 3 Clear-air,
(at ~500 hPa) (cross-nadir) daylight
Uncertainty Δx Δz Δt Number
Layer Orbit Technique Conditions
(RMS) (km) (km) (h) of sats
TIR spectroscopy
LEO 30 nmol · mol–1 50 3 4 3 Clear-air
Troposphere (cross-nadir)
(at ~500 hPa) TIR spectroscopy
GEO 50 nmol · mol–1 50 4 0.5 6 Clear-air
(cross-nadir)
Uncertainty Δx Δz Δt Number
Layer Orbit Technique Conditions
(RMS) (km) (km) (h) of sats
TIR spectroscopy
LEO 30 nmol · mol–1 50 4 4 3 Clear-air
Troposphere (cross-nadir)
(at ~500 hPa) TIR spectroscopy
GEO 50 nmol · mol–1 50 5 0.5 6 Clear-air
(cross-nadir)
CHAPTER 5. SPACE-BASED OBSERVATION OF GEOPHYSICAL VARIABLES 185
Uncertainty Δx Δz Δt Number
Layer Orbit Technique Conditions
(RMS) (km) (km) (h) of sats
Troposphere LEO IR spectroscopy 10 nmol · mol–1 50 4 4 3 Clear-air
(at ~500 hPa) (cross-nadir)
Uncertainty Δx Δz Δt Number
Layer Orbit Technique Conditions
(RMS) (km) (km) (h) of sats
Troposphere LEO IR spectroscopy 10 nmol · mol–1 50 4 4 3 Clear-air
(at ~500 hPa) (cross-nadir)
Uncertainty Δx Δz Δt Number
Layer Orbit Technique Conditions
(RMS) (km) (km) (h) of sats
Troposphere LEO UV spectroscopy 25 nmol · mol–1 50 3 8 3 Clear-air,
(at ~500 hPa) (cross-nadir) daylight
Uncertainty Δx Δz Δt Number
Layer Orbit Technique Conditions
(RMS) (km) (km) (h) of sats
Troposphere LEO SWIR spectroscopy 10 nmol · mol–1 50 4 8 3 Clear-air,
(at ~500 hPa) (cross-nadir) daylight
Uncertainty Δx Δz Δt Number
Layer Orbit Technique Conditions
(RMS) (km) (km) (h) of sats
Troposphere LEO UV spectroscopy 40 nmol · mol–1 50 5 8 3 Clear-air,
(at ~500 hPa) (cross-nadir) daylight
Uncertainty Δx Δz Δt Number
Layer Orbit Technique Conditions
(RMS) (km) (km) (h) of sats
Troposphere LEO IR spectroscopy 10 nmol · mol–1 50 4 4 3 Clear-air
(at ~500 hPa) (cross-nadir)
Uncertainty Δx Δz Δt Number
Layer Orbit Technique Conditions
(RMS) (km) (km) (h) of sats
Troposphere LEO SWIR spectroscopy 20 nmol · mol–1 50 3 8 3 Clear-air,
(at ~500 hPa) (cross-nadir) daylight
Uncertainty Δx Δz Δt Number
Layer Orbit Technique Conditions
(RMS) (km) (km) (h) of sats
Troposphere LEO SWIR spectroscopy 10 nmol · mol–1 100 3 8 3 Clear-air,
(at ~500 hPa) (cross-nadir) daylight
Uncertainty Δx Δz Δt Number
Layer Orbit Technique Conditions
(RMS) (km) (km) (h) of sats
Troposphere LEO TIR spectroscopy 10 nmol · mol–1 50 3 4 3 Clear-air
(at ~500 hPa) (cross-nadir)
Uncertainty Δx Δz Δt Number
Layer Orbit Technique Conditions
(RMS) (km) (km) (h) of sats
Troposphere LEO SW spectroscopy 5 nmol · mol–1 10 3 8 3 Clear-air,
(at ~500 hPa) (cross-nadir) daylight
Uncertainty Δx Δz Δt Number
Layer Orbit Technique Conditions
(RMS) (km) (km) (h) of sats
Stratosphere Sub-mm
LEO 30 nmol · mol–1 300 2 72 1 –
(at ~30 hPa) spectroscopy (limb)
Uncertainty Δx Δz Δt Number
Layer Orbit Technique Conditions
(RMS) (km) (km) (h) of sats
Stratosphere Sub-mm
LEO 15 nmol · mol–1 300 2 72 1 –
(at ~30 hPa) spectroscopy (limb)
190 GUIDE TO INSTRUMENTS AND METHODS OF OBSERVATION - VOLUME IV
Uncertainty Δx Δz Δt Number
Layer Orbit Technique Conditions
(RMS) (km) (km) (h) of sats
Troposphere LEO IR spectroscopy 30 nmol · mol–1 50 3 4 3 Clear-air
(at ~500 hPa) (cross-nadir)
Uncertainty Δx Δz Δt Number
Layer Orbit Technique Conditions
(RMS) (km) (km) (h) of sats
Troposphere LEO SWIR spectroscopy 10 nmol · mol–1 50 3 8 3 Clear-air,
(at ~500 hPa) (cross-nadir) daylight
Uncertainty Δx Δz Δt Number
Layer Orbit Technique Conditions
(RMS) (km) (km) (h) of sats
TIR spectroscopy
LEO 30 nmol · mol–1 50 3 4 3 Clear-air
Troposphere (cross-nadir)
(at ~500 hPa) TIR spectroscopy
GEO 50 nmol · mol–1 50 4 0.5 1 Clear-air
(cross-nadir)
Uncertainty Δx Δz Δt Number
Layer Orbit Technique Conditions
(RMS) (km) (km) (h) of sats
Troposphere LEO SW spectroscopy 20 nmol · mol–1 50 3 8 3 Clear-air,
(at ~500 hPa) (cross-nadir) daylight
Uncertainty Δx Δz Δt Number
Layer Orbit Technique Conditions
(RMS) (km) (km) (h) of sats
Troposphere LEO SW spectroscopy 20 nmol · mol–1 50 3 8 3 Clear-air,
(at ~500 hPa) (cross-nadir) daylight
Uncertainty Δx Δz Δt Number
Layer Orbit Technique Conditions
(RMS) (km) (km) (h) of sats
Sub-mm spectroscopy
LEO 30 nmol · mol–1 300 2 72 1 –
Stratosphere (limb)
(at ~30 hPa) FIR spectroscopy
LEO 20 nmol · mol–1 300 2 72 1 –
(limb)
CHAPTER 5. SPACE-BASED OBSERVATION OF GEOPHYSICAL VARIABLES 193
Table 5.A.7.23. Estimated potential quality of product “Peroxyacetyl nitrate” (by 2020)
Uncertainty Δx Δz Δt Number
Layer Orbit Technique Conditions
(RMS) (km) (km) (h) of sats
TIR spectroscopy
LEO 30 nmol · mol–1 50 3 4 3 Clear-air
Troposphere (cross-nadir)
(at ~500 hPa) TIR spectroscopy
GEO 40 nmol · mol–1 50 4 0.5 6 Clear-air
(cross-nadir)
Uncertainty Δx Δz Δt Number
Layer Orbit Technique Conditions
(RMS) (km) (km) (h) of sats
SW spectroscopy 0.30/0.80
LEO 50 4 8 3 Daylight
(cross-nadir) FAR/HR
SW spectroscopy 0.40/0.70
GEO 50 4 8 6 Daylight
(cross-nadir) FAR/HR
IR spectroscopy 0.25/0.80
LEO 300 2 72 1 –
(limb) FAR/HR
Uncertainty Δx Δz Δt Number
Layer Orbit Technique Conditions
(RMS) (km) (km) (h) of sats
TIR spectroscopy
LEO 25 nmol · mol–1 50 6 4 3 –
(cross-nadir)
TIR spectroscopy
LEO 15 nmol · mol–1 300 2 72 1 –
(limb)
194 GUIDE TO INSTRUMENTS AND METHODS OF OBSERVATION - VOLUME IV
Uncertainty Δx Δz Δt Number
Layer Orbit Technique Conditions
(RMS) (km) (km) (h) of sats
Troposphere LEO UV spectroscopy 20 nmol · mol–1 50 3 8 3 Clear-air,
(at ~500 hPa) (cross-nadir) daylight
8. SPACE WEATHER
Ionospheric total electron content Electron density Magnetic field Electric field
Uncertainty Δx Δz Δt Number
Layer Orbit Technique Conditions
(RMS) (km) (km) (h) of sats
GNSS radio- 90–800 km
LEO 5% 300 3 12 12
occultation altitude
Table 5.A.8.2. Estimated potential quality of product “Electron density” (by 2020)
Uncertainty Δx Δz Δt Number
Layer Orbit Technique Conditions
(RMS) (km) (km) (h) of sats
90–800 km
Ionosphere LEO GNSS radio-occultation 10% 300 10 12 12
altitude
Table 5.A.8.3. Estimated potential quality of product “Magnetic field” (by 2020)
Uncertainty Δx Δz Δt Number
Layer Orbit Technique Conditions
(RMS) (km) (km) (h) of sats
Limited to
LEO Magnetometry 0.3 nT 100 – 240 1
satellite orbit
Magnetosphere
Limited to
GEO Magnetometry 1 nT 100 – 0.25 6
satellite orbit
Table 5.A.8.4. Estimated potential quality of product “Electric field” (by 2020)
Uncertainty Δx Δz Δt Number
Layer Orbit Technique Conditions
(RMS) (km) (km) (h) of sats
Limited to
LEO Ion drift 10 mV·m–1 100 – 240 1
satellite orbit
Magnetosphere
Limited to
GEO Ion drift 10 mV·m–1 100 – 0.25 6
satellite orbit
CHAPTER 6. CALIBRATION AND VALIDATION
6.1.1 Introduction
Calibration is the process of quantitatively defining the satellite instrument response to known,
controlled signal inputs.1 The calibration information is contained in a calibration formula
or in calibration coefficients that are then used to convert the instrument output (measured
in “counts”, or, previously, “analogue signals”) into physical units (for example, radiance
values). Instrument calibration is critical for any higher-level data processing, especially for
deriving quantitative products or when data from different instruments need to be merged
(such as for composite imagery). For climate applications, the requirement for accurate
calibration is particularly stringent 2 since detection of small trends over long periods requires
the ability to compare different instruments flown on different satellites at different times.
Building homogeneous climate data records is contingent on very stable calibration and error
characterization.
Five calibration domains should be generally considered: radiometric, spectral, spatial, temporal,
and polarization. A complete calibration record should include estimates of uncertainties in
calibration parameters. Satellite instrument calibration should take into account all phases of an
instrument’s lifetime: from design and pre-launch phases to post-launch and on-orbit operations.
The response of an instrument to signal input, that is, the relationship between the irradiance
the instrument is exposed to and the numerical value assigned to the measurement (in physical
units, for example, W m–2) depends on several elements, such as:
(a) The viewing geometry, shielding effects, stray light, and antenna pattern;
(c) Filter optics, as well as the possible contamination and stability of the filter;
1
From the Committee on Earth Observation Satellites (CEOS) Working Group on Calibration and Validation.
(The terms defined in the present volume differ in some instances from those defined in JCGM, 2012.)
2
See, for example, Ohring (2007).
3
For guidance on reference standards, see, for example, Fox (2010).
CHAPTER 6. CALIBRATION AND VALIDATION 197
(d) The temperature of all parts of the instrument, including the front-end optics, detector and
back-end electronics (focal plane electronics, preamplifier, etc.);
All of these elements help to determine the spectral response function and the point spread
function that characterize the instrument from a radiometric and geometric viewpoint,
respectively. They must be modelled before launch and monitored in flight by a set of on-board
internal measurements (the housekeeping system). The instrument model and the housekeeping
system are useful for understanding the status of the instrument and its trend as well as for
predicting and correcting biases. However, it is generally not possible to analytically describe the
exact variation of the instrument response resulting from these factors. Reference measurements
are mandatory to characterize the actual calibration.
On-board calibration involves monitoring the instrument performance (and stability) while
in orbit. It is performed using reference targets (such as black bodies in the infrared, solar
diffusers, and lamp line sources in short wave) for passive instruments, or by internal calibration
systems (such as gain monitors) for active instruments. Some heritage instruments have been in
operation without adequate means of on-board calibration, such as the AVHRR, which provides
long-term observations in the visible and near-infrared regions. Other means of calibration
(for example, vicarious, intercalibration) need to be used for characterizing such instruments.
The accuracy of in-flight instrument calibration is a function of the stability of the on-board
calibration systems throughout the instrument’s lifetime. Therefore, the calibration itself must be
regularly checked by intercalibration against highly accurate references.
In the case of infrared instruments, if the radiometer detectors are assumed to have linear
response, the output voltage is given as:
V = α R + V0
where R is the input radiance, α is the radiometer responsivity and V0 is the system offset.
Calibration consists of determining α and V0, which is accomplished by exposing the radiometer
to at least two reference targets with significantly different brightness temperatures.
For infrared and microwave instruments, one reference target is deep space, at a temperature
of 2.725 K. Direct viewing of deep space is not always possible for instruments on a satellite
platform. For instance, pushbroom instruments constantly pointing to the Earth’s surface need
to be equipped with a sub-reflector to supply the deep space view at intervals. A second target
is usually a well-characterized source with temperature in the medium to upper dynamic range,
often a black body, which is ideally traceable to the International System of Units (SI), that is, to a
radiance scale provided by a national metrology institute.
If the instrument response is not linear across the dynamic range, this needs to be accounted
for in the pre-launch instrument characterization, for example by using a quadratic function,
or through linearization in different parts of the dynamic range and the possible addition of a
second black body kept at a different temperature.
198 GUIDE TO INSTRUMENTS AND METHODS OF OBSERVATION - VOLUME IV
For ultraviolet, visible and near-infrared instruments, on-board calibration is more challenging
since it is affected by many factors. At the low-signal extreme, deep space is a useful reference,
provided that disrupting effects (for example, reflections from other parts of the satellite) are
avoided. At the high-signal end, an absolute source is generally replaced by solar diffusers that
provide a relatively stable reference. The moon also may be used as a reference target, with the
advantage that it can be viewed without an attenuator; however, it must be used in conjunction
with an accurate model of the moon’s brightness. Neither the solar diffuser nor the moon
provides an absolute calibration. Another system often used is a bench of lamp line sources of
well-controlled intensity. Spectrally-dependent polarization effects induced by the reflecting
surfaces of the instrument optics also need to be taken into account.
Another problem with on-board calibration is that often the instrument structure does not allow
illumination of the full primary optics with reference sources. For example, a spin-stabilized
radiometer in geostationary Earth orbit uses an internal black body requiring a model of the
contributions of the telescope and foreoptics to the background radiation. Often, the reference
source only illuminates a fraction of the total instrument optics and, therefore, is more used for
stability monitoring than for absolute calibration.
Vicarious calibration can involve different kinds of targets: polar ice fields as a black body for
microwave radiometers; snow fields, sunglint, homogeneous desert areas, and deep convective
cloud tops for the upper end of the visible dynamic range; cloud-free ocean surface for dark
targets in the visible spectrum; cube-corner reflectors for synthetic aperture radars; the rainforest
as a black body for radar scatterometers; etc. Calibration field sites equipped with in situ
observations are used for the calibration of high spatial resolution space-based instruments.
During initial payload commissioning or at regular intervals, aircraft overflights of a target area
synchronous with the satellite overpass offer additional vicarious calibration data.
The intercalibration of satellite instruments involves relating the measurements of one instrument
to those of another. This is done for the dual purposes of:
(b) Merging the data from several instruments to generate consistent time-series.
The intercalibration of instruments operated during the same period requires careful collocation
wherein instrument outputs are compared when the instruments are viewing the same Earth
scenes, at the same times and from the same viewing angles. As part of the International
Satellite Cloud Climatology Project of the World Climate Research Programme, simultaneous
observations from collocations between GEO imagers and an LEO reference imager have been
performed on a monthly basis for almost 30 years as a means to normalize GEO satellite imagery.
More recently, the Global Space-based Inter-calibration System (GSICS) has developed an
operational methodology for such intercalibrations, specifically for simultaneous collocated
CHAPTER 6. CALIBRATION AND VALIDATION 199
(a) Different viewing geometries (with regard to both the instrument scan angle and the solar
position);
(b) Different atmospheric states in the line of sight, including aerosols and clouds;
It should be noted that simultaneous observations between two Sun-synchronous satellites can
only occur at the intersections of their orbital planes, which are always located at a given local
solar time and at a given, generally high north or south, latitude.4
The type of calibration information available depends on the processing level and on the
instrument considered. Each instrument has its own operating mode and calibration cycle, which
includes regular measurements of calibration targets each time a certain number of observations
are performed. For instance, the table in this section indicates the calibration cycles of the
AMSU‑A, the Microwave Humidity Sounder (MHS) and the High-resolution Infrared Sounder 4
(HIRS/4).
Number of cold 2 (deep space ~2.73 K) 4 (deep space ~2.73 K) 56 (deep space ~2.73 K)
target views
An important step in the pre-processing from Level 0 to Level 1b data (see the present volume,
Chapter 2, 2.3.2.6) is to extract the calibration information in the form of warm/cold view counts
and then to compute the resulting calibration coefficients in accordance with the calibration
4
For a 98° inclination, the crossing latitude is above 70° when the ECT of the two orbits differ by less than 8 h, and
only drops significantly when the ECT difference increases towards 12 h.
200 GUIDE TO INSTRUMENTS AND METHODS OF OBSERVATION - VOLUME IV
model (such as a linear or quadratic calibration function, or a lookup table) defined by the
satellite operator for that particular instrument. This provides the operational calibration for that
instrument.
For applications requiring high accuracy and consistency among different instrument data
records, a correction can be applied on top of the operational calibration to take into account
the latest results of the intercalibration activities. Such corrections are provided by GSICS. The
corrected calibration coefficients may be included in the Level 1.b/Level 1.5 data formats as
additional calibration information.
While intercalibration can ensure consistency between satellite instruments, it does not
necessarily provide traceability to SI unless a reference instrument in orbit is SI-traceable. There
are major challenges to achieving SI traceability in orbit as most sensors degrade physically
during and after launch. Achieving SI traceability poses instrument design challenges and
remains a research topic for all but a few measurement types.
The Climate Absolute Radiance and Refractivity Observatory mission proposed by the United
States National Research Council consists of a highly accurate infrared interferometer with
a high-emissivity reference black body using multiple phase-change cells for SI-traceable
thermometer calibration, an ultraviolet, visible and near-infrared spectrometer calibrated by
Sun and moon views, a cryogenically cooled active cavity radiometer, and radio occultation
measurements. This suite of instruments is intended to provide fully traceable measurements of
the entire Earth-emitted and ‑reflected solar spectrum. Implementing and maintaining such a
mission would provide an anchor point in support of the calibration and traceability of the whole
fleet of operational radiometers.
For measurement traceability, one should take advantage of instruments that do not depend on
radiometric calibration, such as radio occultation and Sun or star occultation sensors (see the
present volume, Chapter 2, 2.2.4.3 and 2.2.5.1).
Validation is the process of assessing, by independent means, the quality of the data products
derived from satellite instrument measurements. 5 Product validation should be performed
by product developers, downstream of instrument calibration, and should be documented in
instrument-specific product validation plans. Guidelines for documenting product quality are
provided in the Quality Assurance Framework for Earth Observation (Fox, 2010).
Geophysical products are generated from satellite data (often radiance measurements) by
applying an algorithm that is either physically or empirically based. Comparing the retrieved
products and their trends with in situ observations or model outputs is an important part of
the process to assess and document the reliability of given retrieval algorithms and define their
domain of applicability.
If a particular trend is detected, it may relate to the instrument’s performance; a careful analysis
of the satellite instrument’s calibration and environmental data must be performed before any
empirical correction can be applied.
For many products, validation is a complex problem since the comparison between products
derived from satellite measurements and independent reference products often from in situ
measurements is subject to several errors: (i) an inherent satellite-derived product error, (ii) the
5
From the CEOS Working Group on Calibration and Validation.
CHAPTER 6. CALIBRATION AND VALIDATION 201
error in the reference data, and (iii) the error introduced by the comparison methodology, often
due to non-collocation in time and space. In general, different measurement techniques measure
different things: a satellite observation usually refers to a relatively large area (the instantaneous
field of view) and nearly-instantaneous measurements (within milliseconds); ground (in situ)
measurements are generally very local and integrated over a relatively long time. Surface-
based remote-sensing usually provides information representative of the atmospheric column.
Comparison of the different types of measurements requires downscaling or upscaling methods
that can introduce spatially or temporally dependent errors.
It should be noted that for certain satellite products, independent validation measurements may
not exist, and validation can only be performed by evaluating the impact of the product when
used in an application (for example, when assimilated in an NWP model).
The validation of satellite-derived products should follow defined best-practice and variable-
dependent protocols, such as those developed by the CEOS Working Group on Calibration and
Validation. The validation of satellite-derived parameters and products can be carried out using
the following sources:
Experience shows that the results of direct validation are less significant for some satellite-
derived products than for others. Since validation tests combine the effects of different error
sources (satellite product, ground measurement, comparison method), the error due to the
6
https://w ww.ghrsst.org/products-and-services/product-validation/.
202 GUIDE TO INSTRUMENTS AND METHODS OF OBSERVATION - VOLUME IV
satellite product itself may be difficult to single out. For certain geophysical variables, ground
measurements may be rather inaccurate. For others, the comparison method may depend too
much on the observation environment.
One option for evaluating a product for a particular application is thus to assess its impact on
the application skill. In this case, the evaluation reflects the quality of the product combined
with the ability of the application to use it. For example, NWP models showed marginal impact
for a couple of decades from atmospheric temperature–humidity soundings; this changed to a
significant positive impact only when direct radiance assimilation was introduced. An opposite
example is the assimilation of cloud-motion winds, which exhibited a strong positive impact at
first, although the initial validation exercises were disappointing.
In summary, validation requires rigorous analysis of all error sources and of all steps in the
comparison method. If the analysis shows that the error of the satellite product cannot be singled
out, performing an impact study is a remaining validation mechanism.
CHAPTER 7. CROSS-CUTTING ISSUES
A critical issue for sustaining space-based Earth observations is whether the radio-frequency
spectrum in the microwave range (1 to 300 GHz and above) remains available. This is important
for:
(a) Passive observations of the Earth surface (in atmospheric windows) and atmospheric gases
in absorption bands;
(b) Active observations with radar (altimeters, scatterometers, synthetic aperture radars);
The use of the radio-frequency spectrum is coordinated at the global level by the International
Telecommunication Union (ITU). Radio Regulations are adopted by ITU Members at the
World Radiocommunication Conferences (WRC) every four years. ITU regulates the allocation
of radio-frequency bands to the different applications known as services, such as fixed and
mobile telecommunications, broadband mobile applications, radio navigation, ground-based
radars, short-range devices and electronic news gathering. Earth observation applications are
identified by ITU as two particular services: the Earth exploration-satellite service (EESS) and
Meteorological satellite (Metsat) service. While some bands are allocated to a service on an
exclusive basis, most bands are allocated to several services with certain conditions (such as the
limitation of the number, the emitting power and the geographical distribution of sources) that
aim at avoiding harmful interference.
With the rapid expansion of the telecommunication sector and its increasing spectrum needs,
the protection of frequencies required for EESS and Metsat has become very critical. Concerns
include:
(a) Interference from uncontrolled emissions within exclusive EESS or Metsat bands, or to out-
of-band emissions from nearby frequency bands;
(b) Sharing bands under conditions that are not stringent enough to guarantee reliable
protection;
(c) The desire of other services to expand to bands formerly allocated to EESS or Metsat;
(d) The need of EESS or Metsat to use new bands arising due to the evolving remote-sensing
technology (for example, microwave above 300 GHz), growing data rates or expanding
telecommunication bandwidths.
In addressing these issues, it should be noted that frequencies used for passive measurements
cannot be selected in any part of the spectrum: they are determined by the physics and must
correspond to either absorption peaks of atmospheric components or window channels.
Natural emissions are extremely weak compared to most artificial sources, and hence are easily
corrupted. Therefore, passive radiometric bands must be considered as a natural heritage to be
preserved.
The spectral microwave range used for Earth observation stretches from ~1.4 GHz (for example,
for ocean salinity) to ~2 500 GHz and beyond. The most critical issues are for frequencies below
204 GUIDE TO INSTRUMENTS AND METHODS OF OBSERVATION - VOLUME IV
300 GHz. The use of frequencies above ~300 GHz is still emerging; moreover, since the water
vapour continuum prevents viewing the lower troposphere, most of the instruments operating
at those frequencies are designed for atmospheric chemistry and exploit limb viewing, which
makes them less prone to interferences from ground sources. The ITU has identified a limited
number of bands allocated to EESS, of which active uses are either prohibited or limited. As the
radio-frequency spectrum becomes more crowded and users have a need for higher data rates,
there is increasing pressure on higher frequencies, leading the ITU to share EESS bands with
active services. Only a few narrowbands are assigned to EESS on an exclusive basis ensuring
reliable legal protection. This has the following effects:
(a) The position of the allocated microwave channels often does not coincide with the
sensitivity peak for the needed geophysical variable or with a spectral area free of
contamination from other effects;
(b) The protected bandwidth may be so narrow that the signal-to-noise ratio is poor; this
may force consideration of an unprotected band where a wider bandwidth is available,
accepting the risk of interference.
Unfortunately, the pressure from other users of the spectrum, including commercial and mobile
services, is continuously increasing; specialized groups from WMO, the Coordination Group for
Meteorological Satellites, and space agencies must continuously monitor the situation at each
update of the ITU Regulations.
The problem of frequency protection also holds for active sensing (altimeters, scatterometers
or synthetic aperture radars). For some applications of radar backscattering, like precipitation
measurement, the sensing frequency has to be chosen in relation with the target properties. In
other cases, like altimetry or synthetic aperture radar imagery, it is not very selective and there is
some flexibility to find a frequency allocation in any of the L, S, C, X, Ku, K, Ka, V or W bands (see
the present volume, Chapter 2, Table 2.8 for definitions).
(a) Higher cost for a ground receiving station that works with low signals;
(b) Higher cost because the insufficient bandwidth available in one band for the data rate to
be handled forces relocation to higher-frequency bands that require more challenging
technology and antenna pointing;
(c) Ultimately, more difficulty to secure a frequency, particularly for real-time transmission;
fewer frequencies are available, and sometimes this provokes interferences between
satellites of the same family that are in orbit simultaneously.
In any event, frequency protection is difficult to guarantee, and users are experiencing problems,
especially in industrialized areas. The table in this section provides the frequency bands allocated
to data transmission to and from meteorological satellites (from WMO/ITU, 2008). It also takes
into account the band 7 850–7 900 MHz, which was added at WRC‑12.
CHAPTER 7. CROSS-CUTTING ISSUES 205
The Coordination Group for Meteorological Satellites operates through working groups
dedicated to: (i) satellite systems and telecommunications, (ii) satellite products, (iii) continuity
and contingency planning, and (iv) global data dissemination. Together with WMO, CGMS
launched major collaboration initiatives including the Global Space-based Inter-calibration
System (GSICS), the Sustained and Coordinated Processing of Environmental Satellite Data for
Climate Monitoring project (SCOPE-CM), and the Virtual Laboratory for Education and Training
in Satellite Meteorology. CGMS works in partnership with several international science groups,
is active on a continuous basis, and helps to organize working group meetings at two-year
intervals, including for:
(a) The International TIROS Operational Vertical Sounder (TOVS) Working Group;
The Committee on Earth Observation Satellites (CEOS) was established in 1984 by the Group
of Seven (most industrialized countries of the world). Unlike CGMS, CEOS first focused on land
observation satellites (initially Landsat, then also the Satellite pour l’Observation de la Terre
(SPOT) and the Indian Remote-sensing Satellite (IRS)), and specifically on new technologies
such as synthetic aperture radars. Later, the scope of CEOS was extended to all Earth observation
programmes but with less emphasis on meteorology until a new interest arose for climate issues.
An important activity since its establishment has been calibration and validation through the
Working Group on Calibration and Validation. Other CEOS working groups are the Working
Group on Information Systems and Services, the Working Group on Capacity Building and Data
Democracy, and the new CEOS/CGMS Working Group on Climate.
The coordination activity of CEOS is now articulated around the concept of constellations, to
share experience in the development of algorithms, standardize data products and formats,
exchange information regarding the calibration and validation of measurements, facilitate
the timely exchange of and access to data products from existing and planned missions, and
facilitate the planning of new missions. There are currently seven virtual constellations:
(d) Precipitation;
In the early days of satellite meteorology, the life cycle of a satellite mission was only a few years.
Mission definition, satellite design, system development and science demonstration happened in
one or two years each; the exploitation period was short and the ground segment often under-
dimensioned.
Phase C: Detailed design and development and testing of all systems (including the ground
segment) and subsystems. This phase is the longest, requiring about five years;
Phase D: Integration of all subsystems, testing of the whole satellite and launch campaign.
This is often accomplished in one year.
Adding some time for the decision-making and approval processes, and for in-orbit
commissioning, the duration of the development phase is of the order of 15 years.
The organization of a space programme involves many participants: the applications community,
scientific institutes, space agencies focused on research and development, industry, and
governments with their industrial policy and budget constraints. In the case of Earth observation
programmes with worldwide scope, there must be as much coordination as possible with
international partners, which may further complicate the decision-making process.
The continuity of space-based observations has been a critical requirement for the
meteorological satellite constellation in geostationary orbit ever since nowcasting and severe
weather forecasting, including tropical cyclone warning, began relying on satellite monitoring.
The operational continuity of GEO imagery entails round-the-clock operation, high availability,
near-real-time data dissemination, and long-term continuity guaranteed by a robust programme
that includes provisions for in-orbit backup. When the polar-orbiting constellation was
established, and numerical weather prediction models were increasingly relying on satellite
sounding (infrared, microwave and radio-occultation) and other key satellite observations such
as ocean surface winds, a similar requirement for operational continuity was applied to the
morning and afternoon satellites that became the core meteorological constellation in polar
orbit. A baseline configuration of the operational space-based observing system was defined and
undertaken by the satellite operators contributing to WMO programmes.
While satellite operators committed to give their best efforts to maintain the geostationary
and polar Sun-synchronous constellations, CGMS developed a Global Contingency Plan (see
section 7.2.1 and box below) providing a technical and legal framework for contingency
measures to be taken on a “help your neighbour” basis in case of deficiency of one of the
elements of the operational configuration.
For geostationary satellites, contingency support can be provided if the number of satellites is
sufficiently high and their nominal positions are appropriately spaced along the Equator (see,
for instance, the present volume, Chapter 4, Figure 4.1). Moving a satellite from one longitude
to another requires little fuel if implemented at a slow pace. As per the contingency plan, several
satellites are maintained in backup positions and it is possible to relocate a satellite to fill a gap
through a manoeuvre lasting a few days or weeks depending on the urgency and the fuel
available on board. Several examples have taken place: a spare Meteosat satellite was moved
to cover the West Atlantic Ocean when the Geostationary Operational Environmental Satellite
(GOES) system suffered a launch delay in the early 1990s; a spare GOES satellite filled the gap in
the Western Pacific Ocean during the transition from the Geostationary Meteorological Satellite
208 GUIDE TO INSTRUMENTS AND METHODS OF OBSERVATION - VOLUME IV
(GMS) to the Multifunctional Transport Satellite (MTSAT) in the early 2000s; and in the last
two decades the Indian Ocean position has been covered on occasions by GOES and Meteosat
satellites.
For Sun-synchronous satellites, contingency is more complicated. Changing the orbital plane
of a satellite requires a very large amount of fuel and is not envisaged apart from the natural
drift of the orbital plane due to precession or orbit-keeping manoeuvres to correct this drift. The
contingency plan therefore focuses on the availability of backup satellites in each orbital plane
along with regular spacing of the equatorial crossing times of these planes.
When WMO adopted its Vision for the GOS in 2025, the scope of the operational space-based
observing system was extended to climate monitoring, reflecting both the requirement to
monitor the climate on a continuing basis and the maturity of space-based systems evolving
from research and development to an operational status. In response, a new baseline was
subsequently defined and adopted by CGMS that incorporates a number of climate-oriented
missions. Continuity is as crucial for climate monitoring as for operational weather forecasting;
however, the requirements are different because climate monitoring involves different
timescales. First, near-real-time availability and short-term gaps in a daily cycle are not driving
requirements. Second, major importance is attached to long-term continuity and stability of
measurements throughout decades. The Global Climate Observing System (GCOS) Climate
Monitoring Principles require systematic overlap between all consecutive satellites to allow for
intercalibration and traceability. Stability and traceability could also be achieved by maintaining
one highly secured reference mission, with in-orbit backup, that serves as a calibration reference
standard for all the others (as discussed in the present volume, Chapter 6, 6.1.9). Such a provision
should be a major element in the definition of the Architecture for Climate Monitoring from
Space.
The evolving user requirements for satellite data and the dramatic progress of space and remote-
sensing technology call for continuous improvements to satellite systems and instrumentation.
At the same time, the strong pressure on resources stresses the need to seek an optimization
of the global effort to assure the availability of a comprehensive observing system and avoid
unnecessary redundancy beyond the required margins for robustness. Optimization is also
needed in the development, validation and sustained processing of derived products, and
requires data sharing, interoperability and quality assurance. Global coordination under the
auspices of WMO aims at ensuring such optimization under the overall context of the WMO
CHAPTER 7. CROSS-CUTTING ISSUES 209
Integrated Global Observing System, building on the Rolling Review of Requirements, the
high-level guidance provided by the Vision for the GOS, the Statement of Guidance in each
application area, and the Implementation Plan for the Evolution of Global Observing Systems
which consolidates the recommendations addressed to agents implementing observing systems.
A notable initiative is also the Architecture for Climate Monitoring from Space, promoted by
WMO, CEOS and CGMS, which aims to provide an end-to-end response from the space-based
observing system to the climate monitoring requirements.
REFERENCES AND FURTHER READING
Bizzarri, B. 1982: Satellite data for numerical weather prediction. Rivista di Meteorologia Aeronautica,
42(4):369–382.
———, 1986: Basic principles of remote sensing. Proceedings of the Course on Satellite Meteorology and its
Extension to Agriculture (Erice, ed.). EUMETSAT, EUM P03, pp. 1–10.
Bizzarri, B. and C. Tomassini, 1976: Retrieval of information from high-resolution images. Proceedings
of the Symposium on Meteorological Observations from Space, Committee on Space Research,
Philadelphia, pp. 140–144.
Chander, G., T.J. Hewison, N. Fox, X. Wu, X. Xiong and W.J. Blackwell, 2013: Overview of intercalibration of
satellite instruments. IEEE Transactions on Geoscience and Remote Sensing, 51(3):1056–1080.
Committee on Earth Observation Satellites: Mission, Instruments and Measurements database online
(available from http://database.eohandbook.com, updated yearly).
Datla, R.V., J.P. Rice, K.R. Lykke, C. Johnson, J.J. Butler and X. Xiong, 2009: Best Practice Guidelines for
Pre-launch Characterization and Calibration of Instruments for Passive Optical Remote Sensing.
NIST IR 7637, National Institute of Standards and Technology.
European Space Agency: Sharing Earth Observation Resources (available from https://directory.eoportal
.org/web/eoportal/satellite-missions, updated at intervals).
Fox, N., 2010: QA4EO – A Quality Assurance Framework for Earth Observation: A guide to “reference standards”
in support of Quality Assurance requirements of GEO (M.C. Greening, ed.). QA4EO-QAEO-GEN-
DQK-003, Group on Earth Observations (available from http://www.qa4eo.org/docs/QA4EO
-QAEO - GEN-DQK- 003 _v4.0.pdf).
Joint Committee for Guides in Metrology, 2012: International vocabulary of metrology – Basic and general
concepts and associated terms (VIM). JCGM 200:2012.
Klein, M. and A.J. Gasiewski, 2000: Nadir sensitivity of passive millimeter and submillimeter wave
channels to clear air temperature and water vapor variations. Journal of Geophysical Research,
105(D13):17481–17511.
Kramer, H.J., 2002: Observation of the Earth and its Environment – Survey of Missions and Sensors, Springer (CD-
ROM inclusive of survey on airborne sensors and campaigns).
Ohring, G. (ed.), 2007: Achieving Satellite Instrument Calibration for Climate Change (ASIC3). Report of a
workshop organized by the National Oceanic and Atmospheric Administration, National
Institute of Standards and Technology, National Aeronautics and Space Administration,
National Polar-orbiting Operational Environmental Satellite System-integrated Program Office,
and Space Dynamics Laboratory of Utah State University, 16–18 May 2006. Landsdowne, VA.
Parkinson, C.L., A. Ward and M.D. King (eds.), 2006: Earth Science Reference Handbook. NASA Goddard
Space Flight Center.
Tobin, D.C., H.E. Revercomb, R.O. Knuteson, F.A. Best, W.L. Smith, N.N. Ciganovich, R.G. Dedecker,
S. Dutcher, S.D. Ellington, R.K. Garcia, H.B. Howell, D.D. LaPorte, S.A. Mango, T.S. Pagano,
J.K. Taylor, P. van Delst, K.H. Vinson and M.W. Werner, 2006a: Radiometric and spectral
validation of Atmospheric Infrared Sounder observations with the aircraft-based Scanning
High-Resolution Interferometer Sounder. Journal of Geophysical Research: Atmospheres,
111(D09S02).
Tobin, D.C., H.E. Revercomb, C.C. Moeller and T.S. Pagano, 2006b: Use of Atmospheric Infrared Sounder
high-spectral resolution spectra to assess the calibration of Moderate Resolution Imaging
Spectroradiometer on EOS Aqua. Journal of Geophysical Research: Atmospheres, 111(D09S05).
World Meteorological Organization, 2001: Applications with Meteorological Satellites (W.P. Menzel). SAT-28
(WMO/TD-No. 1078). Geneva.
———, 2014: Observing Systems Capability Analysis and Review (OSCAR) database and tool (available from
http://www.wmo-sat.info/oscar/, regularly updated).
World Meteorological Organization/International Telecommunication Union, 2008: Handbook on Use of
Radio Spectrum for Meteorology: Weather, Water and Climate Monitoring and Prediction. Geneva.
For more information, please contact:
public.wmo.int
JN 181782